Characterization of Thiamine Biosynthetic Enzymes and Their

CHARACTERIZATION OF THIAMINE BIOSYNTHETIC ENZYMES AND THEIR

INTEGRATION IN THE METABOLIC NETWORK OF SALMONELLA ENTERICA

LAUREN DISTERHOFT PALMER

(Under the Direction of Diana Downs)

ABSTRACT

Thiamine pyrophosphate is an essential cofactor required for function of many enzymes in central metabolism. Thiamine is made from two independently synthesized moieties, 4-amino-

5-(hydroxymethyl)-2-methylpyrimidine phosphate (HMP-P), and 4-methyl-5-(2-hydroxyethyl)- thiazole phosphate (THZ-P). Previous work in Salmonella enterica and Escherichia coli had identified all thiamine biosynthetic enzymes and reconstituted most of their activities in vitro.

Gaps remained in the mechanistic understanding of thiamine biosynthetic enzymes, including the

HMP-P synthase ThiC, THZ-P biosynthetic enzyme ThiI, and the fungal HMP-P synthase Thi5p.

This thesis work combined physiological and biochemical approaches to better understand how thiamine biosynthetic enzymes work in the context of S. enterica metabolism.

Mutational analysis of the bacterial HMP-P synthase ThiC variants found no correlation between in vivo and in vitro function, suggesting the in vivo growth phenotypes were more sensitive to changes in metabolite levels than the in vitro ThiC assay. Biochemical studies with improved ThiC in vitro assay conditions resulted in the first report of catalytic turnover, and determined that ThiC was inhibited by a number of S-adenosylmethionine metabolites.

Meanwhile, genetic analysis led to a mechanistic proposal for the requirement of the sulfur

trafficking enzyme ThiI in THZ-P biosynthesis. Nutritional analysis of thiI mutant strains identified an alternative sulfur trafficking pathway in S. enterica when oxidized cysteine metabolites were added to the medium. The Saccharomyces cerevisiae enzyme Thi5p was expressed heterologously in S. enterica to probe cellular factors affecting Thi5p activity and metabolic differences between S. cerevisiae and S. enterica. S. cerevisiae Thi5p functioned conditionally in S. enterica, emphasizing that metabolic modules are not always interchangeable and enzymes can be integrated into the metabolic network in unexpected ways. The findings described in this dissertation highlight the idea that enzyme activity depends on the metabolic context of the cell.

INDEX WORDS: thiamine, vitamin B1, metabolic integration, radical SAM, Salmonella enterica, sulfur

CHARACTERIZATION OF THIAMINE BIOSYNTHETIC ENZYMES AND THEIR

INTEGRATION IN THE METABOLIC NETWORK OF SALMONELLA ENTERICA

LAUREN DISTERHOFT PALMER

B.S., University of Washington, 2008

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

2014

Lauren Disterhoft Palmer

CHARACTERIZATION OF THIAMINE BIOSYNTHETIC ENZYMES AND THEIR

INTEGRATION IN THE METABOLIC NETWORK OF SALMONELLA ENTERICA

LAUREN DISTERHOFT PALMER

Major Professor: Diana Downs

Committee: Michael Adams Jorge Escalante-Semerena Mary Ann Moran

Electronic Version Approved:

Julie Coffield Interim Dean of the Graduate School The University of Georgia August 2014

DEDICATION

I dedicate this dissertation to my family and friends. I want to specially dedicate this dissertation to my life partner, David Wilkerson, who always encouraged me to work hard and keep my project moving forward. You supported me on a day-to-day basis, always ready to pick me up if I missed the bus working in lab, and in the larger scheme, enthusiastically embracing any life-changing decision for the sake of my career. To my parents, Cathy Disterhoft and

Dennis Palmer, who were always excited to read my papers, even if they were “a little dense.”

Thank you for keeping us all connected by taking us on many wonderful family trips. My parents taught me the value of hard work, always encouraged me to keep trying, and were always proud of my accomplishments. To my brother, Alex Palmer, who has always been a great friend, I am excited for you to write your own dissertation in a few years. And to the Wilkersons, Millicent,

John and Emily, who also hosted many wonderful family trips and have always been fantastically supportive of me.

ACKNOWLEDGEMENTS

I would first like to thank my thesis advisor, Diana Downs, for fostering a stimulating scientific environment and for helping me develop scientifically and professionally. I have learned so much from Diana, including many wise phrases like “if you don’t have time to do it right, how are you going to have time to do it again?” I will continue to use her phrases over the years, just one of the ways I will carry her mentorship with me as I move forward in my career.

I was lucky to have the opportunity to work with the faculty, staff and students of two great microbiology programs at the University of Wisconsin-Madison and the University of

Georgia. I thank my committee members at UW, including Heidi Goodrich-Blair, Brian Fox, and especially George Reed and Michael Thomas, who were always willing to take extra time to meet with me. I also thank Barny Whitman and Rob Maier and my committee members at UGA,

Michael Adams and Mary Ann Moran, for valuable discussions. Special thanks to Jorge

Escalante-Semerena, who was on my committee at both schools and always offered encouragement and insightful questions.

I also need to thank all past and present members of the Downs and Escalante labs, I appreciate all of the help over the years. Everyone was always willing to set aside what they were doing and help with protocols, presentations or ideas. Special thanks to Alex Tucker and

Jannell Bazurto for their help with this dissertation, in addition to their countless helpful discussions about experimental design, presentations and manuscripts. My fellow Downs lab members are much more than just coworkers to me, and I value our connections fostered through conversations and shared experiences in graduate school, science, and life.

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ...... v

LIST OF TABLES ...... ix

LIST OF FIGURES ...... x

CHAPTER

1 INTRODUCTION AND LITERATURE REVIEW ...... 1

1.1 THIAMINE BIOSYNTHETIC ENZYMES ...... 2

1.2 METABOLIC INTEGRATION ...... 10

1.3 SYNERGY BETWEEN PHYSIOLOGICAL AND BIOCHEMICAL

APPROACHES ...... 17

1.4 DISSERTATION OUTLINE ...... 18

1.5 REFERENCES ...... 20

2 ANALYSIS OF THIC VARIANTS IN THE CONTEXT OF THE METABOLIC

NETWORK OF SALMONELLA ENTERICA ...... 29

2.1 ABSTRACT ...... 30

2.2 INTRODUCTION ...... 30

2.3 MATERIALS AND METHODS ...... 34

2.4 RESULTS AND DISCUSSION ...... 42

2.5 ACKNOWLEDGEMENTS ...... 53

2.6 REFERENCES ...... 53

vii

3 THE THIAMINE BIOSYNTHETIC ENZYME THIC CATALYZES MULTIPLE

TURNOVERS AND IS INHIBITED BY S-ADENOSYLMETHIONINE (ADOMET)

METABOLITES ...... 58

3.1 ABSTRACT ...... 59

3.2 INTRODUCTION ...... 59

3.3 EXPERIMENTAL PROCEDURES ...... 62

3.4 RESULTS AND DISCUSSION ...... 68

3.5 ACKNOWLEDGEMENTS ...... 76

3.6 REFERENCES ...... 78

4 THE RHODANESE DOMAIN OF THII IS BOTH NECESSARY AND

SUFFICIENT FOR SYNTHESIS OF THE THIAZOLE MOIETY OF THIAMINE

IN SALMONELLA ENTERICA ...... 82

4.1 ABSTRACT ...... 83

4.2 INTRODUCTION ...... 83

4.3 MATERIALS AND METHODS ...... 88

4.4 RESULTS AND DISCUSSION ...... 92

4.5 ACKNOWLEDGEMENTS ...... 103

4.6 REFERENCES ...... 104

5 REDUNDANCY IN SULFUR TRAFFICKING TO THIAMINE BIOSYNTHESIS

IN SALMONELLA ENTERICA ...... 107

5.1 ABSTRACT ...... 108

5.2 INTRODUCTION ...... 108

5.3 MATERIALS AND METHODS ...... 110

viii

5.4 RESULTS AND DISCUSSION ...... 115

5.5 ACKNOWLEDGEMENTS ...... 124

5.6 REFERENCES ...... 124

6 SACCHAROMYCES CEREVISIAE THIAMINE PYRIMIDINE SYNTHASE

FUNCTIONS CONDITIONALLY DURING HETEROLOGOUS EXPRESSION IN

SALMONELLA ENTERICA ...... 130

6.1 ABSTRACT ...... 131

6.2 INTRODUCTION ...... 131

6.3 MATERIALS AND METHODS ...... 134

6.4 RESULTS AND DISCUSSION ...... 141

6.5 ACKNOWLEDGEMENTS ...... 156

6.6 REFERENCES ...... 156

7 CONCLUSIONS ...... 161

7.1 CONCLUSIONS ...... 162

7.2 FUTURE DIRECTIONS ...... 167

7.3 REFERENCES ...... 170

APPENDICES

A COENZYME A MAY PROVIDE ROBUSTNESS TO [FE-S] CLUSTER

METABOLISM IN SALMONELLA ENTERICA ...... 175

A.1 INTRODUCTION ...... 175

A.2 MATERIALS AND METHODS ...... 178

A.3 RESULTS AND DISCUSSION ...... 180

A.4 REFERENCES ...... 187

LIST OF TABLES

Page

TABLE 2.1 Strains used in this study ...... 35

TABLE 2.2 Growth reflects thiamine synthesis allowed by thiC alleles ...... 44

TABLE 2.3 ThiC variants have reduced activity ...... 50

TABLE 4.1 Strain and plasmid list ...... 89

TABLE 4.2 ThiI variants that result in a thiamine requirement in vivo ...... 95

TABLE 4.3 Cysteine allows growth of a thiI mutant in minimal medium ...... 102

TABLE 5.1 Strains, plasmids and primers used in this study ...... 112

TABLE 5.2 Expression from the ydjNp is reduced when cystine is present in the medium ...... 118

TABLE 6.1 Bacterial strains ...... 136

TABLE 6.2 Plasmid and primers ...... 137

TABLE 6.3 Thi5p-dependent growth using different carbon sources ...... 142

TABLE 6.4 Alleles isolated that allow Thi5p-dependent growth in minimal glucose medium ..145

TABLE 6.5 Thi5p-dependent growth on glucose in strains disrupted in glycolysis ...... 154

TABLE A.1 Strains used in this study ...... 179

TABLE A.2 Growth of panE strains containing “adenine-sensitive” ThiC variants ...... 181

LIST OF FIGURES

Page

FIGURE 1.1 THZ-P biosynthesis in bacteria ...... 3

FIGURE 1.2 THZ-P biosynthesis in eukaryotes and archaea ...... 5

FIGURE 1.3 HMP-P biosynthesis ...... 7

FIGURE 1.4 Metabolic integration of thiamine biosynthesis in S. enterica ...... 14

FIGURE 2.1 Thiamine and purine biosynthesis in S. enterica ...... 31

FIGURE 2.2 Primary sequence of ThiC with position of variants ...... 43

FIGURE 2.3 Either methionine or induction of S-adenosylmethionine biosynthesis allowed

growth of strains containing ThiCG273N or ThiCP498L ...... 46

FIGURE 2.4 The growth rate of a strain containing ThiCA527T in a purE mutant background is

increased by exogenous pantothenate ...... 48

FIGURE 2.5 ThiC variants are compromised for HMP formation regardless of substrate

concentration ...... 52

FIGURE 3.1 ThiC reaction ...... 60

FIGURE 3.2 ThiC undergoes steady-state turnover ...... 70

FIGURE 3.3 Metabolite inhibitors of ThiC activity ...... 72

FIGURE 3.4 SAH inhibits ThiC competitively with respect to AdoMet ...... 73

FIGURE 3.5 Cooperative inhibition by 5’-DOA and Met ...... 75

FIGURE 3.6 Adenosine is uncompetitive with AdoMet inhibiting ThiC ...... 77

FIGURE 4.1 Thiamine biosynthesis in S. enterica ...... 84

FIGURE 4.2 Sulfur mobilization by ThiI in 4-thiouridine biosynthesis ...... 86

FIGURE 4.3 Structural domains of S. enterica ThiI ...... 91

FIGURE 4.4 ThiI variants defective in 4-thiouridine biosynthesis complement a thiI mutant

strain ...... 93

FIGURE 4.5 The rhodanese domain of ThiI is sufficient for thiamine-independent growth ...... 97

FIGURE 4.6 Proposed model for ThiI mechanism in thiazole biosynthesis ...... 99

FIGURE 5.1 A ydjN mutation impacts ThiI-independent growth ...... 117

FIGURE 5.2 CdsH is required for cystine or S-sulfocysteine-stimulated growth ...... 120

FIGURE 5.3 Model for ThiI-independent THZ-P biosynthesis ...... 122-123

FIGURE 6.1 Schematic representation of Thi5p-dependent thiamine biosynthesis in S. enterica ...

...... 132

FIGURE 6.2 Carbon sources that support Thi5p-dependent growth in S. enterica ...... 143

FIGURE 6.3 Suppressor mutations allow Thi5p-dependent growth on minimal glucose medium

and form two classes ...... 146

FIGURE 6.4. PEP:glucose phosphotransferase system (PTSGlc) and sugar-phosphate stress

response in S. enterica ...... 147

FIGURE 6.5 Effects of sgrR mutations on transcription from the sgrS promoter ...... 150

FIGURE 6.6 Disruption of PTS allows Thi5p-dependent thiamine biosynthesis on minimal

glucose medium ...... 152

FIGURE A.1 Relevant metabolic pathways in S. enterica ...... 177

FIGURE A.2 The panE yggX strain of S. enterica is auxotrophic for both moieties of thiamine .....

...... 183

FIGURE A.3 The panE yggX strain has slightly increased sensitivity to H2O2 stress ...... 185

CHAPTER 1

INTRODUCTION AND LITERATURE REVIEW

Thiamine pyrophosphate (TPP) is an essential cofactor for all forms of life. TPP allows stabilization of the acyl carbanion (1) and is required for activity of central metabolic enzymes including pyruvate dehydrogenase, α-ketoglutarate dehydrogenase, and transketolase. Thiamine is composed of two independently synthesized moieties, 4-methyl-5-(2-hydroxyethyl)-thiazole phosphate (THZ-P) and 4-amino-5-(hydroxymethyl)-2-methylpyrimidine phosphate (HMP-P).

All organisms require thiamine, but at very low levels: Salmonella enterica strains deficient in thiamine biosynthesis require 10 nM thiamine addition to the growth medium (2) and humans require dietary thiamine with a recommended daily intake of 1.5 mg/day. Thiamine deficiency causes a spectrum of disorders in humans, including beriberi disease or Wernicke–Korsakoff syndrome, whose symptoms include neural or cardiovascular problems such as wasting, numbness, cardiac ataxia, encephalopathy or psychosis.

Thiamine was first identified over 100 years ago when Casimir Funk enriched it from rice bran husks in 1911 as a substance curing a beriberi-like disease in chickens (3); he subsequently coined the term “vital amine” (4) and isolated thiamine in pure form (5). Although thiamine was synthesized and its structure was solved in 1936 (6, 7), identification of the biosynthetic enzymes and reconstitution of their activity was primarily achieved within the past 20 years. Although nearly all thiamine biosynthetic enzymes have been identified and their activity reconstituted in vitro, thiamine biosynthesis remains an active area of research because its enzymes use novel biological chemistry (reviewed in (8) and (9)) and it is used as a system for uncovering metabolic

2 integration (reviewed in (10)). This introduction will describe the thiamine biosynthetic enzymes, focusing on the bacterial and yeast enzymes characterized in this dissertation, discuss metabolic integration of thiamine biosynthesis, and outline the following dissertation.

1.1 THIAMINE BIOSYNTHETIC ENZYMES

1.1.1 THZ-P Biosynthesis

1.1.1.1 Bacterial THZ-P Biosynthesis

THZ-P biosynthesis in bacteria is complex, requiring at least five enzymes. There are two main variants, the first type exemplified by the Enterobacteriaciae Escherichia coli and S. enterica (Fig. 1.1A) and the second by the Firmicute Bacillus subtilis (Fig. 1.1B). In both systems, THZ-P synthase ThiG combines dehydroglycine, deoxy-D-xylulose 5-phosphate (DXP, synthesized by Dxs) and the sulfur from ThiS. The systems differ in the product of the ThiG reaction, their mechanisms of dehydroglycine formation, and their sulfur mobilization schemes.

In E. coli and S. enterica, ThiG produces THZ-P. In B. subtilis, ThiG produces the THZ-P tautomer, rather than THZ-P itself (11); TenI is required for aromatization to THZ-P (12).

Dehydroglycine is produced by the tyrosine lyase ThiH in E. coli/S. enterica, and by the glycine oxidase ThiO in B. subtilis. ThiH is a member of the radical S-adenosylmethionine

(SAM or AdoMet) superfamily of enzymes. Radical SAM enzymes contain an oxygen-labile

2+ [4Fe-4S] clusters coordinated by three cysteines, canonically in a CX3CX2C motif (13), with the fourth Fe molecule coordinated by SAM. Once reduced, the [4Fe-4S]+ cluster reductively cleaves SAM to methionine and the 5’-deoxyadenosyl radical, which then initiates radical catalysis. ThiH is found only in facultative and obligate anaerobes. ThiO, on the other hand, likely relies on oxygen for its activity (14) and is found in obligate aerobes.

A HO ThiH HOOC NH Tyr

HOOC NH2 O OH COOH OP ThiG N S DXP OH

O O ThiF O O ThiF OP THZ-P ThiS OH ThiS O-AMP ThiS S ThiS S ATP S ThiF S ThiI IscS Cys ThiI S SH

B ThiO HOOC NH2 HOOC NH Gly

O OH COOH COOH OP ThiGN TenI N DXP OH S S

O ThiFO ThiF O OP OP THZ-P ThiS OH ThiS O-AMP ThiS SH ATP IscS S IscS SH SH Cys FIGURE 1.1 THZ-P biosynthesis in bacteria. A schematic of THZ-P synthesis in E. coli/S. enterica (A), B. subtilis (B). Abbreviations: Gly, glycine; DXP, deoxy-D-xylulose 5-phosphate; THZ-P, 4-methyl-5-(2-hydroxyethyl)-thiazole phosphate; Cys, cysteine.

In both systems, sulfur mobilization begins with the cysteine desulfurase IscS in E. coli/S. enterica (15) or an IscS homolog in B. subtilis [Park 2003]. A cysteine residue in IscS accepts the sulfur from cysteine, forming an IscS protein persulfide (16). Meanwhile, ThiF activates the terminal glycine of the small protein ThiS by adenylation. In E. coli/S. enterica, the IscS persulfide is proposed to then donate the sulfur group to the ThiI rhodanese domain, forming protein persulfide (17). The ThiI persulfide is proposed to then donate sulfur to the activated

ThiS, forming a disulfide intermediate between the ThiS-thiocarboxylate and ThiI. A cysteine residue in ThiF then forms a disulfide bond with ThiS-thiocarboxylate, producing the observed

ThiF-ThiS acyldisulfide intermediate (18). In E. coli/S. enterica, it is unclear whether the ThiF-

ThiS acyldisulfide then donates sulfur to the THZ-P synthase ThiG, or must first be reduced to the ThiS-thiocarboxylate (18, 19).

B. subtilis THZ-P biosynthesis also uses an IscS homolog, ThiF and ThiS (20). However, the B. subtilis ThiI homolog is not required for thiamine biosynthesis and lacks the relevant rhodanese domain (17, 21). It is not known whether the B. subtilis donates sulfur to an alternative sulfur carrier or directly donates sulfur to activated ThiS. In either case, sulfur donation to the activated ThiS generates the ThiS thiocarboxylate, which is the sulfur donor to B. subtilis ThiG (22). Open questions in THZ-P biosynthesis include the ultimate sulfur donor in E. coli/S. enterica and how the enzymes’ activity is coordinated for THZ-P synthesis.

1.1.1.2 Eukaryotic and archaeal THZ-P synthase: Thi4p

Although at least five enzymes are required for bacterial THZ-P biosynthesis, in eukaryotes such as Saccharomyces cerevisiae only one gene (THI4) was implicated in THZ-P biosynthesis (23). Purification of Thi4p and analysis of partially active mutants identified nicotinamide adenine dinucleotide and glycine as precursors, and identified many intermediates

COOH H N 2 Gly Unidentified COOH NUDIX COOH O + Thi4p N Hydrolase N ADP-O N S S CONH2 HO OH Thi4p O-ADP OP NH THZ-P O

HN SH Thi4p

FIGURE 1.2 THZ-P biosynthesis in eukaryotes and archaea. A schematic of THZ-P synthesis in eukaryotes and the Archaea. Coloring indicates the fate of the atoms. Abbreviations: Gly, glycine; THZ-P, 4-methyl-5-(2-hydroxyethyl)-thiazole phosphate; Cys, cysteine.

in the reaction (Fig. 1.2) (24-27). The source of sulfur for the THZ-P moiety remained elusive until 2011, when Chatterjee et al., reported that Cys205 of the Thi4p enzyme donates H2S for

THZ-P biosynthesis in an Fe-dependent mechanism, and Thi4p is a suicide enzyme (28). This in vitro evidence was corroborated by the fact that Thi4p produced at native levels was modified by a loss of 34 m/z, corresponding to the molecular weight of H2S. Furthermore, S. cerevisiae produced THZ-P with approximately 1:1 stoichiometry to inactivated Thi4p in vivo (28).

As suggested in an accompanying commentary, inactivated Thi4p may have another function in the cell (29). Thi4p accumulates to up to 1.0-1.5% of the total cellular protein in

Neurospora crassa (30), and is associated with oxidative stress in multiple species (31-33).

Further studies are needed to determine whether Thi4p inactivated for THZ-P biosynthesis helps prevent oxidative stress. This connection raises questions about the evolutionary history of Thi4p as a thiamine biosynthetic enzyme and oxidative stress response enzyme.

1.1.2 HMP-P biosynthesis

1.1.2.1 Bacterial/plant/archaeal HMP-P synthase: ThiC

In bacteria, plants, and archaea, ThiC converts the shared thiamine/purine intermediate 5- aminoimidazole ribotide (AIR) to HMP-P in a complex intramolecular rearrangement (Fig. 1.3).

Although in vivo labeling data identified AIR as the sole carbon source for HMP-P biosynthesis decades ago (34-36), ThiC activity was not reconstituted in vitro until 2008 (37, 38). Like ThiH,

ThiC is a member of the radical SAM superfamily. ThiC, like radical SAM enzymes generally, is extremely sensitive to oxygen. Genetic studies found that the ThiC C-terminal domain was required for HMP-P synthesis during aerobic growth, suggesting the C-terminal domain may serve as “cap” to prevent oxidative damage to the [4Fe-4S] cluster (39). In the reported ThiC

A H2N H2N PO O N N ThiC N PO N HO OH AIR HMP-P

B COOH

H2N

His HN H2N N Thi5p N PO N O HO HMP-P

PO N

PLP

FIGURE 1.3 HMP-P biosynthesis. A schematic of HMP-P synthesis in bacteria, plants and Archaea (A) and fungi (B) is illustrated with the relevant gene products and metabolites. Coloring indicates the fate of carbon atoms from the precursors to HMP-P. Abbreviations: AIR, 5-aminoimidazole ribotide; HMP-P, 4-amino-5-(hydroxymethyl)-2-methylpyrimidine phosphate; His, histidine; PLP, pyridoxal 5’-phosphate.

8 structure the C-terminal domain was disordered (likely due to aerobic conditions), but appeared to localize near the active site of the adjacent monomer in the dimer (38).

ThiC has different properties than many members of the radical SAM superfamily. First, it binds its [4Fe-4S] cluster with a unique CX2CX4C motif (37). Second, in vitro labeling studies by Chatterjee et al. found ThiC uses a novel mechanism including two sequential abstractions of hydrogen atoms from the 5’-deoxyadenosyl radical (40). The complete ThiC mechanism is not yet characterized. Because the radical SAM superfamily is an area of active research, it remains unclear whether ThiC is unique or a founding member of a new class of radical SAM enzymes.

Further study on this enzyme will likely expand understanding of biological chemistry carried out by the radical SAM superfamily.

1.1.2.2 Fungal HMP-P synthase: Thi5p family

Fungi do not encode a ThiC homolog; instead, Thi5p is the aerobic HMP-P synthase and the anaerobic HMP-P synthase remains unidentified. Labeling studies in S. cerevisiae first showed that under aerobic conditions, the carbon in HMP was derived from histidine and pyridoxine or pyridoxal 5’-phosphate (PLP) (41-43), rather than the purine intermediate AIR.

Genetic studies showed that all four Thi5p family homologs in S. cerevisiae (Thi5p (YFL058w),

Thi11p (YJR156c), Thi12p (YNL332w) and Thi13p (YDL244w)) were functionally redundant, i.e. only the quadruply mutant strain was a thiamine auxotroph (44). Other yeast species have between zero and five copies of THI5 family genes (44). By contrast, the anaerobic HMP-P synthase does not seem to use a pyridoxine vitamer as a substrate (44, 45).

Lai et al. reported the only biochemical reconstitution of Thi5p activity in 2012. They reported that Thi5p activity depended on the presence of an Fe cofactor, oxygen, and PLP (9).

Their data showed that in vitro, Thi5p His66 was the substrate histidine, suggesting Thi5p may

9 be a single turnover (suicide) enzyme (9). Another structural study published shortly afterwards confirmed that His66 was required for Thi5p function in vivo in S. cerevisiae and their structural analysis suggested a free histidine molecule would not fit in the active site (46). The idea that

Thi5p is a suicide enzyme raises interesting questions about selective pressures for maintaining

Thi5p for HMP-P biosynthesis. There is evidence of horizontal gene transfer from bacteria to yeast (47), suggesting yeast could potentially acquire ThiC. Because yeast have an alternate anaerobic HMP-P biosynthetic pathway and ThiC is extremely oxygen-sensitive, perhaps the fact that Thi5p relies on oxygen is important for its maintenance.

As discussed above, there is strong evidence to indicate the eukaryotic THZ-P biosynthetic enzyme, Thi4p, is a suicide enzyme (28). Coquille et al. suggest having both THZ-P and HMP-P synthesized by suicide enzymes could provide an alternative mechanism to prevent precursor (e.g. PLP) depletion because yeast lack TPP riboswitches to control thiamine biosynthetic enzyme expression (46). However, although yeast lack TPP riboswitches, THI expression is repressed when thiamine is present in the growth medium, indicating some form of transcriptional regulation exists (reviewed in (48)). Evidence that Thi5p is a suicide enzyme is limited; however, Thi5p family members are among the most highly produced enzymes in the S. cerevisiae proteome in thiamine-free medium (49). Future studies probing Thi5p function in vivo, similar to those conducted with Thi4p, could determine whether suicide enzymes produce both thiamine moieties in yeast.

1.1.3 TPP Synthesis

After de novo biosynthesis of HMP-P, it is phosphorylated to HMP-PP by HMP kinase

(ThiD in E. coli/S. enterica, N-terminal domain of Thi20p in yeast) (50-52). Thiamine synthase

(ThiE in E. coli/S. enterica; N-terminal domain of Thi6p in yeast) condenses THZ-P and HMP-

PP to thiamine monophosphate (TMP) (53, 54). Finally, thiamine phosphate kinase ThiL phosphorylates TMP to the biologically active cofactor, TPP (55). In yeast, TMP is first phosphatased to thiamine by an unknown enzyme and then thiamine is pyrophosphorylated to

TPP by Thi80p (56).

1.2 METABOLIC INTEGRATION

Metabolism is often thought of in terms of conserved metabolic pathways or modules, such as glycolysis or thiamine biosynthesis. Metabolic integration describes how these modules are connected through metabolites and enzymes. These connections include shared metabolites, shared enzymes, and non-enzymatic chemistry, among others. For example, fructose-6- phosphate is an essential precursor for cell wall biosynthesis and is shared by glycolysis and the pentose phosphate pathway. The last five steps of isoleucine and valine biosynthesis share the same enzymes but convert different precursors in parallel pathways. The enzyme ThiI (discussed above and in Chapters 4 and 5) is required for sulfur mobilization to both the thiamine thiazole and the tRNA photosensor 4-thiouridine. Through metabolic integration of these modules, every metabolite in the cell is connected to one another.

Metabolic integration can also be mediated by enzyme inhibition, which can provide a method of metabolic control. For example, amino-4-imidazole carboxamide ribotide (AICAR) is a purine biosynthetic intermediate and a byproduct of histidine biosynthesis, and thus is a shared metabolite. AICAR inhibits enzymes in other pathways, including adenosine deaminase in purine catabolism (57), fructose-1,6-bisphosphatase (FBP) in gluconeogenesis (58), and pantoate

β-alanine ligase in Coenzyme A biosynthesis (59). The enzymes AICAR is known to inhibit use adenosine-containing substrates, suggesting AICAR inhibition could be due to its similarity to

11 other purines. In this way, it is unclear whether some metabolic integration may be an unintended side effect of metabolism or a selected mode of metabolic control.

1.2.1 Conservation of metabolic integration

Metabolism is inherently integrated because there are only 12 essential precursors for all biomass production, and all can be generated in glycolysis, the pentose phosphate pathway and the TCA cycle. Metabolism seems to have evolved to take the shortest biochemical path between these essential precursors, and each carbon source is catabolized by the shortest path to an essential precursor (60). Metabolic integration is conserved in terms of linking glycolysis, the pentose phosphate pathway and the TCA cycle. Beyond the integration of central carbon metabolism, metabolic integration is variable. For example, by simply comparing substrates used for HMP-P biosynthesis we see differences in metabolic integration between fungi v. bacteria, plants, and archaea. In fungi that use Thi5p, HMP-P biosynthesis is linked to histidine and PLP biosynthesis, whereas in organisms that use ThiC, it is linked to purine biosynthesis. These differences may be maintained by unidentified selective pressures and reflect fundamental metabolic differences between organisms.

1.2.2 Metabolic integration of thiamine

Vitamins and cofactors mediate metabolic integration as products of biosynthesis branching from central metabolism and as required cofactors for many metabolic pathways, including central carbon metabolism. For example, there are TPP dependent enzymes in the pentose phosphate pathway (transketolase), TCA cycle (pyruvate dehydrogenase and α- ketoglutarate dehydrogenase), and thiamine biosynthesis itself (deoxyxylulose-5-phosphate synthase (61)). The centrality of vitamin cofactors to metabolism is likely due to their evolutionary history. In 1976, White first proposed that vitamin cofactors helped early ribozymes

12 mediate chemical reactions (62); this hypothesis was proven possible in a 2013 study identifying a ribozyme that used thiamine to decarboxylate a pyruvate-like suicide substrate (63). Perhaps due to its ancient evolutionary history, thiamine biosynthesis is highly integrated in the metabolic network.

Thiamine biosynthesis in S. enterica is a model system for dissection of metabolic integration. This model system relies on the absolute but low requirement for thiamine (10 nM), which allows sensitive detection of changes in flux by a growth/no growth phenotypic output.

Due to the genetic tractability of S. enterica, suppressor analysis has been used to identify metabolic integration by selecting for thiamine-independent growth under conditions when a given strain cannot synthesize its own thiamine. As a low flux biosynthetic system, suppressor analysis of thiamine biosynthesis allows detection of subtle changes in flux.

In addition to its utility as a model system, metabolic integration of thiamine could have industrial implications. Thiamine is widely used in human nutrition products, which has greatly reduced incidence of thiamine deficiency diseases such as beriberi. But current industrial synthesis of thiamine uses expensive starting materials or generates highly carcinogenic reactants

(discussed in (64)). Chemical synthesis of thiamine is generally difficult, and each complex synthesis scheme results in different byproducts that can affect the final purity and quality of the thiamine produced (64). Process research groups are still working to improve chemical thiamine synthesis, and even with recent improvements, current syntheses yield only 65%-70% (65). The cost and difficulty of chemical thiamine synthesis suggest bioproduction could be feasible.

Understanding cellular factors that affect thiamine biosynthesis could aid in production of an industrially relevant thiamine-producing strain of bacteria or yeast.

1.2.2.1 Thiamine connections to purine biosynthesis and alternative purine biosynthesis

As mentioned previously, the first five steps in HMP-P biosynthesis are shared with purine biosynthesis (66); this branched pathway was the first identified instance of metabolic integration of thiamine biosynthesis (Fig. 1.4). A S. enterica strain lacking first step of the shared

HMP-P/thiamine biosynthetic pathway, the amidophosphoribosyltransferase PurF, is a purine auxotroph and a conditional thiamine auxotroph (67). Genetic analysis of the conditional thiamine auxotrophy of the purF mutant strain identified metabolic integration with other areas of metabolism (reviewed in (68, 69). The purF mutant strain is deficient in the metabolite phosphoribosylamine (PRA), the product of the PurF reaction. In permissive growth conditions, sufficient PRA is generated for thiamine biosynthesis but not purine biosynthesis (67). Genetic screens for thiamine auxotrophs in permissive conditions identified a link between HMP-

P/purine biosynthesis and the oxidative pentose phosphate pathway (70). Suppressor analysis of the purF mutant strain in non-permissive growth conditions identified mutations in the ridA gene, formerly yjgF (71). Investigation of the role of RidA in thiamine biosynthesis led to identification of its biochemical activity, which has since defined a new form of endogenous metabolic stress (71-78).

Selections of suppressor mutations that allow derivatives of the purF to grow in non- permissive conditions have also identified links between HMP-P/purine biosynthesis and tryptophan biosynthesis, histidine biosynthesis, and the central metabolic precursor phosphoribosylpyrophosphate (PRPP) (10, 72, 79-81). Many of these mechanisms are predicted to increase the intracellular pool ribose-5-phosphate. Ribose-5-phosphate appears to condense non-enzymatically with ammonia in vivo to generate sufficient PRA for HMP-P biosynthesis

(10, 79), emphasizing the contributions of non-enzymatic chemistry to the metabolic network.

Tryptophan PrsA Biosynthesis TrpD + IlvA - RidA

PurF PurD PurT PurG PurI PurK PurE PurC PurB PurH PRA AIR AICAR IMP

PPP ThiC Histidine CoA Biosynthesis [Fe-S] metabolism HMP-P SAM Methionine

THZ-P ThiIFSGH, IscS TPP

FIGURE 1.4 Metabolic integration of thiamine biosynthesis in S. enterica. A schematic of thiamine synthesis is illustrated with the gene products involved in purine biosynthesis shown near the relevant position in the pathway. Metabolic processes that impact ThiC activity in vivo are depicted: solid lines indicate direct interactions and dotted lines indicate indirect interactions. Abbreviations: PRA, phosphoribosylamine; AIR, 5-aminoimidazole ribotide; AICAR, amino-4- imidazole carboxamide ribotide; IMP, inosine monophosphate; HMP-P, 4-amino-5- hydroxymethyl-2-methylpyrimidine phosphate; CoA, coenzyme A; SAM, S- adenosylmethionine; THZ-P, 4-methyl-5-β-hydroxyethylthiazole phosphate; TPP, thiamine pyrophosphate.

The study of metabolic integration of thiamine has identified new activities of known enzymes

(including TrpDE) and identified a new metabolic stress (RidA).

1.2.2.2 Thiamine biosynthesis and iron sulfur cluster metabolism

YggX is a small protein that binds Fe2+ in vitro and is required for protection from oxidative stress (82, 83). In strains defective in producing the YggX, genetic analysis identified six gene products (IscA, GshA, RseC, ApbC, ApbE, and CyaY) that were conditionally required for thiamine biosynthesis in S. enterica (84-86). These proteins were required for biosynthesis of both thiamine moieties (THZ-P and HMP-P) (84, 87); CyaY was required when both ApbC and

YggX are lacking and excess Fe was provided in the medium (86). The defect in THZ-P biosynthesis was found to be caused by the oxygen-labile [4Fe-4S] cluster in ThiH (84, 88), suggesting an HMP-P biosynthetic enzyme also contained an oxygen-labile [Fe-S] cluster.

Further analysis determined that [Fe-S] cluster metabolism affected the conversion of

AIR to HMP-P, suggesting ThiC could also use an oxygen-labile [Fe-S] cluster for catalysis (39).

Methionine was reported to be required for HMP-P biosynthesis, but labeling studies in S. enterica showed that methionine was not an HMP-P precursor (66, 89). Methionine is a precursor to SAM, and these findings together suggested that ThiC was a radical SAM enzyme.

The idea that ThiC could be a radical SAM enzyme led to successful implementation of the first reported reconstitution of ThiC activity in vitro (37). The identification of ThiC as a radical SAM enzyme exemplifies how understanding metabolic integration can inform biochemical studies of difficult-to-reconstitute enzymes with novel chemistry.

1.2.2.3 Thiamine biosynthesis and coenzyme A

Mutations reducing coenzyme A (CoA) levels in the cell caused a conditional HMP auxotrophy, identifying a link between HMP-P biosynthesis and CoA metabolism in S. enterica.

CoA is a required coenzyme that serves as an acyl group carrier used by central metabolic enzymes in the TCA cycle, fatty acid biosynthesis and the synthesis of amino acids. Strains lacking the ketopantoate reductase PanE have a 90% reduction in CoA compared to the wild- type strain (90). The branched chain amino acid biosynthetic enzyme acetohydroxyacid isomeroreductase (IlvC) has ketopantoate reductase activity (91) that produces the remaining

10% of CoA, which is sufficient to prevent an auxotrophy (90). This example of metabolic integration provides metabolic redundancy in CoA biosynthesis.

Lesions in panE result in an HMP auxotrophy when flux through the shared purine/HMP-

P biosynthetic pathway is compromised by eliminating or inhibiting PurF (90, 92, 93). In the presence of purines, expression of purF is repressed and the PurF enzyme is allosterically inhibited (94-96). CoA specifically affects the ThiC-catalyzed conversion of AIR to HMP-P in vivo (97), but does not affect ThiC activity in vitro (98). The mechanism of the CoA effect on

ThiC is unknown (and discussed further in Appendix A).

1.2.2.4 AICAR effects on thiamine biosynthesis

As discussed above, the purine biosynthetic intermediate and histidine biosynthetic byproduct AICAR inhibits enzymes in diverse metabolic processes. A purH mutant strain accumulates AICAR and has reduced HMP-P biosynthesis in S. enterica, which complicated early genetic studies characterizing the integration of the purine and thiamine biosynthetic pathway (66, 99). AICAR does not inhibit ThiC activity in vitro (98), and analysis thus far suggests its effect is due to multiple, additive causes. AICAR inhibition of pantoate β-alanine ligase and the resultant defect in CoA biosynthesis likely contributes to ThiC inhibition in vivo

(59). The purH mutant thiamine auxotrophy was overcome by addition of pantothenate, a CoA precursor, and/or methionine (59). The rescue by pantothenate and methionine was additive,

17 suggesting the purH strain was inhibited in another metabolic process. Methionine could potentially stimulate ThiC activity in a purH strain through SAM biosynthesis, and future studies may test that hypothesis. Interestingly, AICAR can also serve as a precursor for alternative AIR synthesis (100). The purine biosynthetic enzyme PurC can convert AICAR to AIR directly (J.

Bazurto and D. Downs, personal communication), which is the only described alternative thiamine biosynthetic pathway that does not generate PRA.

1.3 SYNERGY BETWEEN PHYSIOLOGICAL AND BIOCHEMICAL APPROACHES

Although study of metabolic integration is necessarily rooted in physiological studies, it can also provide insight into biochemical mechanisms. As an example, physiological and biochemical studies of the biotin synthase BioB, a radical SAM enzyme, led to molecular explanations of metabolic integration phenotypes and improvement of the reconstituted assay. E. coli strains lacking pfs, which encodes methylthioadenosine nucleosidase (MTAN), were partial biotin auxotrophs (101). MTAN was known as a shared enzyme for degradation of methylthioadenosine and S-adenosylhomocysteine, SAM-derived products of spermidine biosynthesis and SAM-dependent methylation (tangentially, MTAN is also important for production of autoinducer-2 for quorum sensing). Another group reported that 5’- deoxyadenosine, the product of radical SAM cleavage, inhibited BioB activity in vitro (102).

These reports informed a study by Choi-Rhee et al. reporting that MTAN degradation of 5’- deoxyadenosine prevented physiologically relevant BioB inhibition when exogenous 5’- deoxyadenosine was added to the culture medium (103).

Later biochemical studies of BioB used MTAN to degrade 5’-deoxyadenosine in situ, which contributed to the first report of BioB catalytic turnover in vitro (104). The same biochemical study of BioB reported potent inhibition by S-adenosylhomocysteine (104). Thus,

18 the original finding that a pfs mutant required biotin for optimal growth was likely due to elevated levels of both S-adenosylhomocysteine and 5’-deoxyadenosine. In this way, biochemical and physiological studies combined provide more insight into metabolic integration than either alone. The work in Chapter 3 was built on the finding that SAM metabolites could inhibit radical SAM enzymes, which was key to successfully implementing a catalytic ThiC assay.

1.4 DISSERTATION OUTLINE

The goal of this dissertation was to understand how thiamine biosynthetic enzymes work within the context of the metabolic network, using biochemical and physiological approaches synergistically. The remaining chapters are summarized below.

Chapter 2 describes a physiological and biochemical study of the bacterial HMP-P synthase ThiC in S. enterica. Here I characterized thiC mutants that are conditional thiamine auxotrophs, probing the effects of perturbations in flux through the purine biosynthetic operon, and addition of methionine or pantothenate. This analysis suggested that the reconstituted ThiC assay was insufficiently sensitive to allow mechanistic characterization of subtle metabolic changes affecting HMP-P biosynthesis in the cell.

Chapter 3 describes an improved ThiC assay and the first report of its catalytic turnover in vitro. Inhibitor analysis found that ThiC was inhibited by many SAM metabolites, including what may be general radical SAM enzyme inhibitors such as S-adenosylhomocysteine and 5’- deoxyadenosine. ThiC was also inhibited by adenosine and methylthioadenosine, and is the only reported radical SAM enzyme inhibited by these metabolites.

In Chapter 4, the mechanistic function of ThiI in THZ-P biosynthesis was elucidated.

Mutational analysis led to the finding that the ThiI C-terminal rhodanese domain was necessary

19 and sufficient for sulfur mobilization to THZ-P biosynthesis. Bioinformatic analysis found that less than one quarter of the ThiI proteins in annotated genomes had the C-terminal rhodanese domain, suggesting that the majority of “ThiI” proteins were misannotated as thiamine biosynthetic enzymes.

Chapter 5 further investigated the finding, reported in Chapter 4, that addition of exogenous cysteine could rescue a thiI mutant. Genetic analysis found that YdjN, a transporter of oxidized cysteine metabolites, was required for the cysteine rescue. This led to the finding that oxidized cysteine products, rather than cysteine itself rescued the thiI strain. This rescue also required the cysteine desulfhydrase CdsH, suggesting in vivo production of sulfide could generate persulfides capable of replacing the ThiI persulfide for sulfur mobilization to thiamine.

Chapter 6 describes characterization of the yeast HMP-P synthase Thi5p in the context of the S. enterica metabolic network. This study was initiated before Thi5p activity was reconstituted in vitro. I found that expression of THI5 complemented an S. enterica thiC mutant strain conditionally suggesting there are fundamental differences between the metabolic context in S. enterica and S. cerevisiae. Thi5p functioned on certain carbon sources, such as ribose and mannose, but not others, such as glucose. This work identified the first reported constitutive activators of the sugar-phosphate stress response. Although the mechanism of glucose-inhibition of Thi5p is not yet understood, genetic analysis suggests direct inhibition by glycolytic intermediates is unlikely.

Chapter 7 summarizes conclusions from this work and discusses future directions.

Appendix A describes studies to understand the effect CoA has on the ThiC reaction in the cell. Observations reported here suggest that reduced CoA can also compromise the THZ-P

20 biosynthetic enzyme ThiH under certain conditions, suggesting CoA may be important for in vivo function of radical SAM enzymes or [Fe-S] cluster enzymes generally.

1.5 REFERENCES

1. Hong J, Sun S, Derrick T, Larive C, Schowen KB, Schowen RL. 1998. Transition-state theoretical interpretation of the catalytic power of pyruvate decarboxylases: the roles of static and dynamical considerations. Biochim. Biophys. Acta. 1385:187-200.

2. Enos-Berlage JL, Downs DM. 1997. Mutations in sdh (succinate dehydrogenase genes) alter the thiamine requirement of Salmonella typhimurium. J. Bacteriol. 179:3989-3996.

3. Funk C. 1911. On the chemical nature of the substance which cures polyneuritis in birds induced by a diet of polished rice. J. Physiol. 43:395-400.

4. Funk C. 1912. The etiology of the deficiency diseases. J. State Med.:341-368.

5. Funk C. 1913. Studies on beri-beri: VII. Chemistry of the vitamine-fraction from yeast and rice-polishings. J. Physiol. 46:173-179.

6. Williams RR. 1936. Structure of vitamin B1. J. Am. Chem. Soc. 58:1063-1064.

7. Williams RR, Cline JK. 1936. Synthesis of vitamin B1. J. Am. Chem. Soc. 58:1504-1505.

8. Begley TP, Chatterjee A, Hanes JW, Hazra A, Ealick SE. 2008. Cofactor biosynthesis-- still yielding fascinating new biological chemistry. Curr. Opin. Chem. Biol. 12:118-125.

9. Lai R-Y, Huang S, Fenwick MK, Hazra A, Zhang Y, Rajashankar K, Philmus B, Kinsland C, Sanders JM, Ealick SE, Begley TP. 2012. Thiamin pyrimidine biosynthesis in Candida albicans: a remarkable reaction between histidine and pyridoxal phosphate. J. Am. Chem. Soc. 134:9157-9159.

10. Koenigsknecht MJ, Fenlon LA, Downs DM. 2010. Phosphoribosylpyrophosphate synthetase (PrsA) variants alter cellular pools of ribose 5-phosphate and influence thiamine synthesis in Salmonella enterica. Microbiology 156:950-959.

11. Hazra A, Chatterjee A, Begley TP. 2009. Biosynthesis of the thiamin thiazole in Bacillus subtilis: identification of the product of the thiazole synthase-catalyzed reaction. J. Am. Chem. Soc. 131:3225-3229.

12. Hazra AB, Han Y, Chatterjee A, Zhang Y, Lai RY, Ealick SE, Begley TP. 2011. A missing enzyme in thiamin thiazole biosynthesis: identification of TenI as a thiazole tautomerase. J. Am. Chem. Soc. 133:9311-9319.

13. Sofia HJ, Chen G, Hetzler BG, Reyes-Spindola JF, Miller NE. 2001. Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with

radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res. 29:1097-1106.

14. Settembre EC, Dorrestein PC, Park JH, Augustine AM, Begley TP, Ealick SE. 2003. Structural and mechanistic studies on ThiO, a glycine oxidase essential for thiamin biosynthesis in Bacillus subtilis. Biochemistry 42:2971-2981.

15. Lauhon CT, Kambampati R. 2000. The iscS gene in Escherichia coli is required for the biosynthesis of 4-thiouridine, thiamin and NAD. J. Biol. Chem. 275:20096-20103.

16. Kambampati R, Lauhon CT. 2000. Evidence for the transfer of sulfane sulfur from IscS to ThiI during the in vitro biosynthesis of 4-thiouridine in Escherichia coli tRNA. J. Biol. Chem. 275:10727-10730.

17. Martinez-Gomez NC, Palmer LD, Vivas E, Roach PL, Downs DM. 2011. The rhodanese domain of ThiI is both necessary and sufficient for synthesis of the thiazole moiety of thiamine in Salmonella enterica. J. Bacteriol. 193:4582-4587.

18. Xi J, Ge Y, Kinsland C, McLafferty FW, Begley TP. 2001. Biosynthesis of the thiazole moiety of thiamin in Escherichia coli: identification of an acyldisulfide-linked protein-- protein conjugate that is functionally analogous to the ubiquitin/E1 complex. Proc. Natl. Acad. Sci. U. S. A. 98:8513-8518.

19. Taylor SV, Kelleher NL, Kinsland C, Chiu HJ, Costello CK, Backstrom AD, McLafferty FW, Begley TP. 1998. Thiamin biosynthesis in Escherichia coli - Identification of ThiS thiocarboxylate as the immediate sulfur donor in the thiazole formation. J. Biol. Chem. 273:16555-16560.

20. Park JH, Dorrestein PC, Zhai H, Kinsland C, McLafferty FW, Begley TP. 2003. Biosynthesis of the thiazole moiety of thiamin pyrophosphate (vitamin B1). Biochemistry 42:12430-12438.

21. Rajakovich LJ, Tomlinson J, Dos Santos PC. 2012. Functional analysis of Bacillus subtilis genes involved in the biosynthesis of 4-thiouridine in tRNA. J. Bacteriol. 194:4933-4940.

22. Dorrestein PC, Zhai H, McLafferty FW, Begley TP. 2004. The biosynthesis of the thiazole phosphate moiety of thiamin: the sulfur transfer mediated by the sulfur carrier protein ThiS. Chem. Biol. 11:1373-1381.

23. Praekelt UM, Byrne KL, Meacock PA. 1994. Regulation of THI4 (MOL1), a thiamine- biosynthetic gene of Saccharomyces cerevisiae. Yeast 10:481-490.

24. Chatterjee A, Jurgenson CT, Schroeder FC, Ealick SE, Begley TP. 2007. Biosynthesis of thiamin thiazole in eukaryotes: Conversion of NAD to an advanced intermediate. J. Am. Chem. Soc. 129:2914-2922.

25. Chatterjee A, Jurgenson CT, Schroeder FC, Ealick SE, Begley TP. 2006. Thiamin biosynthesis in eukaryotes: Characterization of the enzyme-bound product of thiazole synthase from Saccharomyces cerevisiae and its implications in thiazole biosynthesis. J. Am. Chem. Soc. 128:7158-7159.

26. Jurgenson CT, Chatterjee A, Begley TP, Ealick SE. 2006. Structural insights into the function of the thiamin biosynthetic enzyme Thi4 from Saccharomyces cerevisiae. Biochemistry 45:11061-11070.

27. Chatterjee A, Schroeder FC, Jurgenson CT, Ealick SE, Begley TP. 2008. Biosynthesis of the thiamin-thiazole in eukaryotes: Identification of a thiazole tautomer intermediate. J. Am. Chem. Soc. 130:11394-11398.

28. Chatterjee A, Abeydeera ND, Bale S, Pai PJ, Dorrestein PC, Russell DH, Ealick SE, Begley TP. 2011. Saccharomyces cerevisiae THI4p is a suicide thiamine thiazole synthase. Nature 478:542-546.

29. Roach P. 2011. Biochemistry: Suicide of a protein. Nature 478:463-464.

30. Faou P, Tropschug M. 2004. Neurospora crassa CyPBP37: a cytosolic stress protein that is able to replace yeast Thi4p function in the synthesis of vitamin B1. J. Mol. Biol. 344:1147-1157.

31. Machado CR, Praekelt UM, de Oliveira RC, Barbosa AC, Byrne KL, Meacock PA, Menck CF. 1997. Dual role for the yeast THI4 gene in thiamine biosynthesis and DNA damage tolerance. J. Mol. Biol. 273:114-121.

32. Ruiz-Roldan C, Puerto-Galan L, Roa J, Castro A, Di Pietro A, Roncero MI, Hera C. 2008. The Fusarium oxysporum sti35 gene functions in thiamine biosynthesis and oxidative stress response. Fung. Gen. Biol. 45:6-16.

33. Medina-Silva R, Barros MP, Galhardo RS, Netto LE, Colepicolo P, Menck CF. 2006. Heat stress promotes mitochondrial instability and oxidative responses in yeast deficient in thiazole biosynthesis. Res. Microbiol. 157:275-281.

34. Estramareix B, Lesieur M. 1969. Biosynthesis of the pyrimidine portion of thiamine: source of carbons 2 and 4 in Salmonella typhimurium. Biochim. Biophys. Acta 192:375- 377.

35. Estramareix B, Therisod M. 1984. Biosynthesis of thiamine: 5-aminoimidazole ribotide as the precursor of all the carbon-atoms of the pyrimidine moiety. J. Am. Chem. Soc. 106:3857-3860.

36. Estramareix B, David S. 1990. Conversion of 5-aminoimidazole ribotide to the pyrimidine of thiamin in enterobacteria: study of the pathway with specifically labeled samples of riboside. Biochim. Biophys. Acta 1035:154-160.

37. Martinez-Gomez NC, Downs DM. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

38. Chatterjee A, Li Y, Zhang Y, Grove TL, Lee M, Krebs C, Booker SJ, Begley TP, Ealick SE. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

39. Dougherty MJ, Downs DM. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

40. Chatterjee A, Hazra AB, Abdelwahed S, Hilmey DG, Begley TP. 2010. A "radical dance" in thiamin biosynthesis: mechanistic analysis of the bacterial hydroxymethylpyrimidine phosphate synthase. Angew. Chem. Int. Ed. Eng. 49:8653- 8656.

41. Tazuya K, Yamada K, Kumaoka H. 1993. Pyridoxine is a precursor of the pyrimidine moiety of thiamin in Saccharomyces cerevisiae. Biochem. Mol. Biol. Int. 30:893-899.

42. Tazuya K, Yamada K, Kumaoka H. 1989. Incorporation of histidine into the pyrimidine moiety of thiamin in Saccharomyces cerevisiae. Biochim. Biophys. Acta 990:73-79.

43. Ishida S, Tazuya-Murayama K, Kijima Y, Yamada K. 2008. The direct precursor of the pyrimidine moiety of thiamin is not urocanic acid but histidine in Saccharomyces cerevisiae. J. Nutr. Sci. Vitaminol. 54:7-10.

44. Wightman R, Meacock PA. 2003. The THI5 gene family of Saccharomyces cerevisiae: distribution of homologues among the hemiascomycetes and functional redundancy in the aerobic biosynthesis of thiamin from pyridoxine. Microbiology 149:1447-1460.

45. Tanaka K, Tazuya K, Yamada K, Kumaoka H. 2000. Biosynthesis of thiamin under anaerobic conditions in Saccharomyces cerevisiae. Biol Pharm Bull 23:108-111.

46. Coquille S, Roux C, Fitzpatrick TB, Thore S. 2012. The last piece in the vitamin B1 biosynthesis puzzle: structural and functional insight into yeast 4-amino-5- hydroxymethyl-2-methylpyrimidine phosphate (HMP-P) synthase. J. Biol. Chem. 287:42333-42343.

47. Hall C, Brachat S, Dietrich FS. 2005. Contribution of horizontal gene transfer to the evolution of Saccharomyces cerevisiae. Euk. Cell 4:1102-1115.

48. Hohmann S, Meacock PA. 1998. Thiamin metabolism and thiamin diphosphate- dependent enzymes in the yeast Saccharomyces cerevisiae: genetic regulation. Biochim. Biophys. Acta 1385:201-219.

49. Muller EH, Richards EJ, Norbeck J, Byrne KL, Karlsson KA, Pretorius GH, Meacock PA, Blomberg A, Hohmann S. 1999. Thiamine repression and pyruvate decarboxylase

autoregulation independently control the expression of the Saccharomyces cerevisiae PDC5 gene. FEBS Lett. 449:245-250.

50. Reddick JJ, Kinsland C, Nicewonger R, Christian T, Downs DM, Winkler ME, Begley TP. 1998. Overexpression, purification and characterization of two pyrimidine kinases involved in the biosynthesis of thiamin: 4-amino-5-hydroxymethyl-2-methylpyrimidine kinase and 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate kinase. Tetrahedron 54:15983-15991.

51. Petersen LA, Downs DM. 1997. Identification and characterization of an operon in Salmonella typhimurium involved in thiamine biosynthesis. J. Bacteriol. 179:4894-4900.

52. Kawasaki Y, Onozuka M, Mizote T, Nosaka K. 2005. Biosynthesis of hydroxymethylpyrimidine pyrophosphate in Saccharomyces cerevisiae. Curr. Genet. 47:156-162.

53. Backstrom AD, Mcmordie RAS, Begley TP. 1995. Biosynthesis of thiamin I: the function of the thiE gene product. J. Am. Chem. Soc. 117:2351-2352.

54. Nosaka K, Nishimura H, Kawasaki Y, Tsujihara T, Iwashima A. 1994. Isolation and characterization of the THI6 gene encoding a bifunctional thiamin-phosphate pyrophosphorylase/hydroxyethylthiazole kinase from Saccharomyces cerevisiae. J. Biol. Chem. 269:30510-30516.

55. Webb E, Downs D. 1997. Characterization of thiL, encoding thiamin-monophosphate kinase, in Salmonella typhimurium. J. Biol. Chem. 272:15702-15707.

56. Nosaka K, Kaneko Y, Nishimura H, Iwashima A. 1993. Isolation and characterization of a thiamin pyrophosphokinase gene, THI80, from Saccharomyces cerevisiae. J. Biol. Chem. 268:17440-17447.

57. Kuramitsu HK, Udaka S, Moyed HS. 1964. Induction of inosine 5'-phosphate dehydrogenase and xanthosine 5'-phosphate aminase by ribosyl-4-amino-5- imidazolecarboxamide in purine-requiring mutants of Escherichia coli B. J. Biol. Chem. 239:3425-3430.

58. Dougherty MJ, Boyd JM, Downs DM. 2006. Inhibition of fructose-1,6-bisphosphatase by aminoimidazole carboxamide ribotide prevents growth of Salmonella enterica purH mutants on glycerol. J. Biol. Chem. 281:33892-33899.

59. Bazurto JV, Downs DM. 2014. Amino-4-imidazolecarboxamide ribotide directly inhibits coenzyme A biosynthesis in Salmonella enterica. J. Bacteriol. 196:772-779.

60. Noor E, Eden E, Milo R, Alon U. 2010. Central carbon metabolism as a minimal biochemical walk between precursors for biomass and energy. Mol. Cell 39:809-820.

61. Sprenger GA, Schorken U, Wiegert T, Grolle S, deGraaf AA, Taylor SV, Begley TP, BringerMeyer S, Sahm H. 1997. Identification of a thiamin-dependent synthase in

Escherichia coli required for the formation of the 1-deoxy-D-xylulose 5-phosphate precursor to isoprenoids, thiamin, and pyridoxol. Proc. Nat. Acad. Sci. U. S. A 94:12857- 12862.

62. White HB. 1976. Coenzymes as fossils of an earlier metabolic state. J. Mol. Evol. 7:101- 104.

63. Cernak P, Sen D. 2013. A thiamin-utilizing ribozyme decarboxylates a pyruvate-like substrate. Nat. Chem. 5:971-977.

64. Eggersdorfer M, Laudert D, Letinois U, McClymont T, Medlock J, Netscher T, Bonrath W. 2012. One hundred years of vitamins-a success story of the natural sciences. Angew. Chem. 51:12960-12990.

65. Zhao L, Ma XD, Chen FE. 2012. Development of two scalable syntheses of 4-amino-5- aminomethyl-2-methylpyrimidine: key intermediate for vitamin B1. Org. Process Res. Dev. 16:57-60.

66. Newell PC, Tucker RG. 1968. Biosynthesis of the pyrimidine moiety of thiamine. A new route of pyrimidine biosynthesis involving purine intermediates. Biochem. J. 106:279- 287.

67. Petersen L, Enos-Berlage J, Downs DM. 1996. Genetic analysis of metabolic crosstalk and its impact on thiamine synthesis in Salmonella typhimurium. Genetics 143:37-44.

68. Downs DM. 2006. Understanding microbial metabolism. Annual Rev. Microbiol. 60:533- 559.

69. Koenigsknecht MJ, Downs DM. 2010. Thiamine biosynthesis can be used to dissect metabolic integration. Trends Microbiol. 18:240-247.

70. Enos-Berlage JL, Downs DM. 1996. Involvement of the oxidative pentose phosphate pathway in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 178:1476- 1479.

71. Enos-Berlage JL, Langendorf MJ, Downs DM. 1998. Complex metabolic phenotypes caused by a mutation in yjgF, encoding a member of the highly conserved YER057c/YjgF family of proteins. J. Bacteriol. 180:6519-6528.

72. Browne BA, Ramos AI, Downs DM. 2006. PurF-independent phosphoribosyl amine formation in yjgF mutants of Salmonella enterica utilizes the tryptophan biosynthetic enzyme complex anthranilate synthase-phosphoribosyltransferase. J. Bacteriol. 188:6786- 6792.

73. Lambrecht JA, Browne BA, Downs DM. 2010. Members of the YjgF/YER057c/UK114 family of proteins inhibit phosphoribosylamine synthesis in vitro. J. Biol. Chem. 285:34401-34407.

74. Lambrecht JA, Flynn JM, Downs DM. 2012. Conserved YjgF protein family deaminates reactive enamine/imine intermediates of pyridoxal 5'-phosphate (PLP)-dependent enzyme reactions. J. Biol. Chem. 287:3454-3461.

75. Lambrecht JA, Downs DM. 2013. Anthranilate phosphoribosyl transferase (TrpD) generates phosphoribosylamine for thiamine synthesis from enamines and phosphoribosyl pyrophosphate. ACS Chem. Biol. 8:242-248.

76. Lambrecht JA, Schmitz GE, Downs DM. 2013. RidA proteins prevent metabolic damage inflicted by PLP-dependent dehydratases in all domains of life. mBio 4:e00033-00013.

77. Flynn JM, Christopherson MR, Downs DM. 2013. Decreased coenzyme A levels in ridA mutant strains of Salmonella enterica result from inactivated serine hydroxymethyltransferase. Mol. Microbiol. 89:751-759.

78. Flynn JM, Downs DM. 2013. In the absence of RidA, endogenous 2-aminoacrylate inactivates alanine racemases by modifying the pyridoxal 5'-phosphate cofactor. J. Bacteriol. 195:3603-3609.

79. Ramos I, Vivas EI, Downs DM. 2008. Mutations in the tryptophan operon allow PurF- independent thiamine synthesis by altering flux in vivo. J. Bacteriol. 190:815-822.

80. Ramos I, Downs DM. 2003. Anthranilate synthase can generate sufficient phosphoribosyl amine for thiamine synthesis in Salmonella enterica. J. Bacteriol. 185:5125-5132.

81. Koenigsknecht MJ, Lambrecht JA, Fenlon LA, Downs DM. 2012. Perturbations in histidine biosynthesis uncover robustness in the metabolic network of Salmonella enterica. PLOS One 7:e48207.

82. Gralnick J, Downs D. 2001. Protection from superoxide damage associated with an increased level of the YggX protein in Salmonella enterica. Proc. Natl. Acad. Sci. U. S. A. 98:8030-8035.

83. Cui Q, Thorgersen MP, Westler WM, Markley JL, Downs DM. 2006. Solution structure of YggX: a prokaryotic protein involved in Fe(II) trafficking. Proteins 62:578-586.

84. Skovran E, Downs DM. 2000. Metabolic defects caused by mutations in the isc gene cluster in Salmonella enterica serovar typhimurium: implications for thiamine synthesis. J. Bacteriol. 182:3896-3903.

85. Skovran E, Lauhon CT, Downs DM. 2004. Lack of YggX results in chronic oxidative stress and uncovers subtle defects in Fe-S cluster metabolism in Salmonella enterica. J. Bacteriol. 186:7626-7634.

86. Vivas E, Skovran E, Downs DM. 2006. Salmonella enterica strains lacking the frataxin homolog CyaY show defects in Fe-S cluster metabolism in vivo. J. Bacteriol. 188:1175- 1179.

87. Skovran E, Downs DM. 2003. Lack of the ApbC or ApbE protein results in a defect in Fe-S cluster metabolism in Salmonella enterica serovar Typhimurium. J. Bacteriol. 185:98-106.

88. Gralnick J, Webb E, Beck B, Downs D. 2000. Lesions in gshA (Encoding gamma-L- glutamyl-L-cysteine synthetase) prevent aerobic synthesis of thiamine in Salmonella enterica serovar typhimurium LT2. J. Bacteriol. 182:5180-5187.

89. Newell PC, Tucker RG. 1968. Precursors of the pyrimidine moiety of thiamine. Biochem. J. 106:271-277.

90. Frodyma M, Rubio A, Downs DM. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar typhimurium. J. Bacteriol. 182:236-240.

91. Primerano DA, Burns RO. 1983. Role of acetohydroxy acid isomeroreductase in biosynthesis of pantothenic acid in Salmonella typhimurium. J. Bacteriol. 153:259-269.

92. Downs DM, Petersen L. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.

93. Frodyma ME, Downs D. 1998. ApbA, the ketopantoate reductase enzyme of Salmonella typhimurium is required for the synthesis of thiamine via the alternative pyrimidine biosynthetic pathway. J. Biol. Chem. 273:5572-5576.

94. Rolfes RJ, Zalkin H. 1988. Regulation of Escherichia coli purF. Mutations that define the promoter, operator, and purine repressor gene. J. Biol. Chem. 263:19649-19652.

95. Rolfes RJ, Zalkin H. 1988. Escherichia coli gene purR encoding a repressor protein for purine nucleotide synthesis. Cloning, nucleotide sequence, and interaction with the purF operator. J. Biol. Chem. 263:19653-19661.

96. Zhou G, Smith JL, Zalkin H. 1994. Binding of purine nucleotides to two regulatory sites results in synergistic feedback inhibition of glutamine 5-phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 269:6784-6789.

97. Allen S, Zilles JL, Downs DM. 2002. Metabolic flux in both the purine mononucleotide and histidine biosynthetic pathways can influence synthesis of the hydroxymethyl pyrimidine moiety of thiamine in Salmonella enterica. J. Bacteriol. 184:6130-6137.

98. Palmer LD, Downs DM. 2013. The thiamine biosynthetic enzyme ThiC catalyzes multiple turnovers and is inhibited by S-adenosylmethionine (AdoMet) metabolites. J. Biol. Chem. 288:30693-30699.

99. Yura T. 1956. Evidence of nonidentical alleles in purine requiring mutants of Salmonella typhimurium. Publ. Carnegie Inst. 612:63-75.

100. Bazurto JV, Downs DM. 2011. Plasticity in the purine-thiamine metabolic network of Salmonella. Genetics 187:623-631.

101. Cornell KA, Riscoe MK. 1998. Cloning and expression of Escherichia coli 5'- methylthioadenosine/S-adenosylhomocysteine nucleosidase: identification of the pfs gene product. Biochim. Biophys. Acta 1396:8-14.

102. Tse Sum Bui B, Lotierzo M, Escalettes F, Florentin D, Marquet A. 2004. Further investigation on the turnover of Escherichia coli biotin synthase with dethiobiotin and 9- mercaptodethiobiotin as substrates. Biochemistry 43:16432-16441.

103. Choi-Rhee E, Cronan JE. 2005. A nucleosidase required for in vivo function of the S- adenosyl-L-methionine radical enzyme, biotin synthase. Chem. Biol. 12:589-593.

104. Farrar CE, Siu KK, Howell PL, Jarrett JT. 2010. Biotin synthase exhibits burst kinetics and multiple turnovers in the absence of inhibition by products and product-related biomolecules. Biochemistry 49:9985-9996.

CHAPTER 2

ANALYSIS OF THIC VARIANTS IN THE CONTEXT OF THE

METABOLIC NETWORK OF SALMONELLA ENTERICA1

1 Palmer L.D., M.J. Dougherty, D.M. Downs. 2012. Journal of Bacteriology. 194(22):6088-95. Reprinted here with the permission of the publisher.

2.1 ABSTRACT

In bacteria, the 4-amino-hydroxymethyl-2-methylpyrimidine (HMP) moiety of thiamine is synthesized from 5-aminoimidazole ribotide phosphate (AIR), a branch point metabolite of purine and thiamine biosynthesis. ThiC is a member of the radical S-adenosylmethionine

(AdoMet) superfamily and catalyzes the complex chemical rearrangement of AIR to HMP. As reconstituted in vitro, the ThiC reaction requires AdoMet, AIR, and reductant. This study analyzed variants of ThiC in vivo and in vitro to probe the metabolic network surrounding AIR in

Salmonella enterica. Several variants of ThiC that required metabolic perturbations to function in vivo, were biochemically characterized in vitro. Results herein indicate that the subtleties of the metabolic network have not been captured in the current reconstitution of the ThiC reaction.

2.2 INTRODUCTION

Thiamine pyrophosphate is an essential cofactor used by enzymes in central metabolism such as pyruvate dehydrogenase and α-ketoglutarate dehydrogenase to stabilize the acyl carbanion (22). Thiamine is composed of two independently synthesized moieties, 4-methyl-5-

(2-hydroxyethyl)-thiazole (THZ) and 4-amino-5-(hydroxymethyl)-2-methylpyrimidine (HMP).

In all bacteria characterized to date, the pyrimidine of thiamine is synthesized from a branch point in purine biosynthesis. In a single step ThiC (HMP-P synthase) converts the branch point metabolite, 5-aminoimidazole ribotide phosphate (AIR) to hydroxymethyl pyrimidine phosphate

(10, 27). HMP-P is sequentially phosphorylated and combined with THZ to generate thiamine monophosphate (2, 31) before a final phosphorylation generates the active cofactor (43) (Fig.

2.1).

In vivo, thiamine synthesis is integrated with purine biosynthesis by the shared branch point metabolite AIR. During growth in minimal medium, the majority of flux through AIR is

PurF PurDTGI PurKE PurCBH PRA AIR CAIR IMP H H2N C C N

H2 O N CH C H PO C CH HC CH AdoMet Met HO OH AIR ThiC [Fe-S] ThiC

NH2 CoA Pant H2 C C N C OP

H C CH + CO CH N + HCOOH H HMP-P HMP-P

S OP N THZ-P TPP

FIGURE 2.1 Thiamine and purine biosynthesis in S. enterica. A schematic of thiamine synthesis is illustrated with the gene products involved in purine biosynthesis shown near the relevant position in the pathway. The inset displays the fate of individual atoms in the synthesis of HMP from AIR based on labeling studies (9, 16-18, 26). Metabolic processes that impact ThiC activity in vivo are depicted. Stains lacking PurR will be de-repressed for all purine genes. Abbreviations: AIR, aminoimidazole ribotide; THZ, 4-methyl-5-β-hydroxyethylthiazole phosphate; HMP, 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate, TPP, thiamine pyrophosphate.

32 used for purine biosynthesis and a minor portion is diverted to thiamine biosynthesis.

Nutritional studies suggest a 1:1000 ratio of flux to thiamine and purine synthesis would satisfy both requirements (1). Despite its role as a precursor to HMP-P, the synthesis of AIR is not regulated by thiamine; this fact raises the question of how S. enterica ensures the thiamine requirement of the cell is satisfied under conditions where purine biosynthesis is repressed.

Significantly, the natural gut habitat of Salmonella has excess purines (39). This suggests that during steady state growth in the gut, PurF (amidophosphoribosyltransferase, EC:2.4.2.14) will be allosterically inhibited and the flux through the purine pathway will be low (44).

A number of results have shown that when purine biosynthetic flux is reduced, thiamine synthesis is maintained, suggesting that the metabolites flow to thiamine synthesis prior to purines. Mutants with less than 1% wild-type activity of the purine biosynthetic enzyme PurI

(phosphoribosylaminoimidazole synthetase, EC:6.3.3.1), require purines but not thiamine (1).

Furthermore, several pathways recruited to feed into the purine pathway, by forming either phophoribosyl amine (PRA) or AIR, satisfy the thiamine but not purine requirement of the cell

(4, 15, 24, 30, 32, 33). In other words, in a wild-type cell, the partitioning of flux to thiamine biosynthesis is relatively insensitive to changes in flux to AIR. This type of insensitivity is thought to be characteristic of systems when the substrate concentration is well above the enzyme Km value (5). Thus, the ThiC Km for AIR is expected to be lower than the AIR concentration in the cell, but this hypothesis cannot be addressed with current enzymatic assays of ThiC. It is possible that ThiC variants can be used to probe the AIR branch point, because the sensitivity of the branch point can amplify relatively small changes in enzyme characteristics

(25). It follows that this system could also be used to monitor subtle changes in the metabolic network.

In S. enterica, a number of metabolic perturbations have been shown to impact the conversion of AIR to HMP-P by ThiC. In vivo experiments found that compromised iron-sulfur cluster ([Fe-S]) cluster metabolism reduced ThiC activity, results that contributed to the identification of the [Fe-S] requirement of this protein (13). In other studies, strains with reduced coenzyme A (CoA) levels, and those that accumulated the purine intermediate 5- aminoimidazole-4-carboxamide ribotide (AICAR) have less ThiC-dependent growth in vivo (1,

20). Neither CoA nor AICAR had a demonstrable effect on the ThiC reaction in vitro, raising the question of how the respective perturbations were affecting the activity of the enzyme in vivo (N.

Cecilia Martinez-Gomez and Diana Downs, unpublished).

Work on ThiC in the past five years has established an in vitro assay and begun to define some of its biochemical parameters. ThiC is a dimeric member of the radical SAM superfamily of proteins that use an [Fe-S] cluster and S-adenosylmethionine (SAM or AdoMet) to carry out radical-mediated catalysis (10, 27, 34). In the presence of AdoMet and reductant in vitro, ThiC converts AIR to HMP-P, carbon monoxide, and formate (9). A structure of Caulobacter crescentus ThiC without the Fe-S cluster was solved (10), and modeling both the cluster and the active site has provided insights into the reaction mechanism of this protein. A catalytic mechanism for the conversion of AIR to HMP-P has been proposed based on data that suggested the 5'-deoxyadenosyl radical sequentially abstracts two hydrogen atoms from AIR in the reaction

(9). Despite the recent progress in our understanding of HMP-P synthesis, the ThiC reaction mechanism has not been fully characterized and enzymatic turnover has not yet been achieved.

We suggest that better understanding of the metabolic environment relevant to ThiC in vivo may aid in improving enzyme activity in vitro.

This study was initiated to use our ability to analyze the in vivo system and characteristics of the in vitro ThiC assay to query changes in the cellular environment that are caused by specific metabolic perturbations. Specifically, we sought to use ThiC variants that required perturbations in metabolism to function in vivo as a means to better understand how ThiC is integrated in the metabolic network, and whether the in vitro assay adequately reflects the cellular properties relevant to the function of ThiC.

2.3 MATERIALS AND METHODS

Strains, media and chemicals. The strains used in this study were derived from S. enterica serovar Typhimurium LT2, and are listed in Table 2.1. The indicated thiC alleles and a linked marker (zxx-8039::Tn10d(Tc)) are present in each of four strain backgrounds: wild-type; purE3043; purR3090; purE3043 purR3090. Tn10d(Tc) refers to the transposition-defective mini-

Tn10 described by Way et al. (42).

Rich medium (NB) is Difco nutrient broth (8 g/L) with NaCl (5 g/L). Superbroth (SB) is tryptone (32 g/L), yeast extract (20 g/L), NaCl (5 g/L) with NaOH (0.05 N). Minimal medium is no-carbon essential (NCE) medium supplemented with MgSO4 (1 mM), trace minerals (0.1 X, adapted from Balch et al.(3)), and glucose (11 mM). Difco BiTek agar was added (15 g/L) for solid medium. When present in the medium, supplements were provided at the following concentrations: thiamine, 100 nM; HMP, 100 nM; adenine, 0.4 mM; methionine, 0.3 mM; pantothenate, 50 µM. Antibiotics were added at the following concentrations in rich and minimal media, respectively: ampicillin (Ap), 150 mg/L, 30 mg/L; tetracycline (Tc), 20 mg/L, 10 mg/L; kanamycin (Km), 50 mg/L, 12.5 mg/L. All chemicals were purchased from Sigma-Aldrich, St

Louis, MO.

TABLE 2.1 Strains used in this study.a

thiC Protein b Conditional Mutagen c Wild-type purE purR purE purR Allele Variant Auxotroph WT N/A DM7778 DM13262 DM13850 DM13274 thiC1076 G486D HA N DM4003 DM13264 DM13276 thiC1128 E281K HA Y DM7777 DM13292 DM13851 DM13295 thiC1129 V267M HA Y DM7779 DM13293 DM13852 DM13296 thiC1130 S247F HA N DM7781 DM13263 DM13275 thiC1140 R397H HA N DM13314 DM13265 DM13277 thiC1141 G481S HA N DM13315 DM13286 DM13290 thiC1142 D468N HA N DM13316 DM13266 DM13278 thiC1143 G479R HA N DM13317 DM13287 DM13291 thiC1144 H501Y HA N DM13318 DM13267 DM13279 thiC1145 G472D HA N DM13319 DM13268 DM13280 thiC1146 G92D HA Y DM13320 DM13269 DM13853 DM13281 thiC1147 G273N HA Y DM13321 DM13288 DM13854 DM13297 thiC1158 A527T NTG Y DM13322 DM13294 DM13855 DM13298 thiC1159 D509G NTG Y DM13323 DM13289 DM13856 DM13299 thiC1161 P498L NTG Y DM13324 DM13270 DM13857 DM13282 thiC1163 G355D NTG N DM13325 DM13271 DM13283 thiC1171 R544C NTG N DM13326 DM13272 DM13284 D61N, thiC1172 NTG N DM13327 DM13273 DM13284 G513E

a thiC mutant alleles are listed with the predicted protein variant, mutagen used to generate the allele, and corresponding strain numbers in various genetic backgrounds. b generated by hydroxylamine (HA (23)), or nitrosoguanidine (NTG (12)) mutagenesis. c depicts whether the relevant allele can support thiamine synthesis sufficient for growth under any tested condition, Y, yes and N, no.

Mutant isolation and strain construction. A lysate of P22 (HT int201 (36, 40)) was generated on strain DM851 that had been mutagenized with nitrosoguanidine (NTG) (12). Alternatively, a

P22 lysate grown on strain DM851 was mutagenized by hydroxylamine (HA) (23). DM851 carries zxx-8039::Tn10d (Tc), which is ~80% linked to the thi locus. Each lysate was used to transduce a wild-type strain to TcR. TcR transductants were screened for those that required exogenous thiamine or HMP for growth under one of three conditions: (1) in the presence of exogenous adenine; (2) in a purF mutant background; (3) at increased temperature. After isolation, the thiC alleles were transduced into the appropriate backgrounds by P22. An isogenic pair (thiC+ and thiC mutant alleles) from each transduction was purified to be phage free and saved (8). In the backgrounds where the thiC alleles did not display thiamine auxotrophy, the presence of the mutation was confirmed by backcrosses. Accumulation of the ThiC variant proteins in the wild-type background was verified by Western Blot analysis.

Growth analysis. Cells from overnight cultures in NB medium were pelleted and resuspended in an equal volume of saline (0.85% NaCl), and a 5 µL aliquot was used to inoculate 195 µL of the appropriate minimal medium. Growth was quantified in a 96-well plate using a microplate reader

(model EL808, Bio-tek Instruments). Cell density was measured as optical density (OD) at 650 nm, and growth was reported as a specific growth rate (µ = ln(X/X0)/T).

Expression and purification of Fpr/FldA. Plasmids to express S. enterica Flavoprotein reductase (Fpr) and flavodoxin (FldA) were constructed with pET-28b(+) (Novagen) using the

NheI and XhoI sites. Fpr and FldA were individually expressed in E. coli strain BL21 (DE3)

(Novagen). In each case, a 10-mL overnight culture grown in SB + Ap was used to inoculate two

2.8 L Fernbach flasks, each containing 1.5 L of SB + Ap. The culture was incubated at 37°C with shaking to an OD650 nm of approximately 0.9, IPTG was added to 1 mM and incubation

37 continued overnight at 28°C with shaking. The cells were harvested by centrifugation and resuspended to 30 mL Buffer A (50 mM HEPES, pH 7.5). Lysozyme (30 mg) and DNase (2 mg) were added and the cells were lysed by two passages through a French pressure cell at 1500 psi.

The lysate was clarified by centrifugation at 41,400 X g for 45 min, filtered with a 0.45 µm PES filter (Whatman) and loaded on a Ni-NTA Superflow (Qiagen) column (7 mL). For both Fpr and

FldA preps, the column was washed with 75 mL Buffer A, then 25 mL with 3% Buffer B (50 mM HEPES, pH 7.5, 500 mM imidazole, 500 mM Na2SO4). The proteins were then eluted with a linear gradient (3-100% B) over 10 column volumes. The proteins were concentrated and FAD and FMN were added to 1mM final concentration to Fpr and FldA, respectively. The proteins were then dialyzed into 2 X 1 L Buffer A, then 1 L freezing buffer (50 mM HEPES, pH 7.5, 10% glycerol). The proteins were dropped into liquid nitrogen and the beads were stored at -80°C.

Protein concentration was determined by Pierce 660 (Pierce) assay using BSA as a standard. Fpr concentration was 9.2 mg/mL (0.3 mM) and FldA was 6.7 mg/mL (0.3 mM).

Expression and purification of ThiC. Wild-type and mutant thiC were cloned into pET-28b(+)

(Novagen) using the NheI and XhoI sites. Plasmids expressing wild-type ThiC or the ThiC variants were co-electroporated into E. coli strain BL21AI (Novagen) with the plasmid pDB1282, which expresses Azotobacter vinelandii [Fe-S] cluster-loading genes (iscSUA hscBA fdx orf3 ndK) (11). Two 2.8-L Fernbach flasks with 1.5 L SB + Ap + Km were each inoculated with a 10 mL culture of overnight culture of the appropriate strain. The cultures were incubated at 37°C with shaking to an OD650 of 0.3. Arabinose was added to 0.2% to induce expression of the nif genes, and FeCl3 was added to 100 µM. Incubation continued for 3 h before expression of thiC was induced by adding IPTG to a final concentration of 1 mM, and incubation continued

38 with shaking overnight at 15°C. Cells were harvested by centrifugation at 16,000 x g for 15 min.

Cell pellets were routinely stored at -80°C until use.

All manipulations with ThiC proteins were carried out in a Coy glove box maintained at

<2 ppm oxygen as detected by an oxygen sensor (Coy Laboratory Products, Inc.). The cell pellet from 3 L culture (generally 15-25 g wet cell weight) was resuspended to 30 mL with anoxic

Buffer A (50 mM HEPES pH 7.5, 0.2 M Na2SO4, 12% glycerol). Cells were incubated on ice with lysozyme (4 mg/ml) and DNase (2 mg) for a minimum of 10 min before sonication.

Sonication was performed on ice, with 30 s sonication and 1 min rests for 15 min using a 60

Sonic Dismembrator (Fisher Scientific) emitting 15-25 watts. The cell lysate was clarified by centrifugation at 48,000 x g for 45-60 min in sealed Oakridge tubes outside of the glovebox. The clarified lysate was returned to the glovebox, filtered, and applied to a Ni-NTA Superflow resin column (6 mL) equilibrated in Buffer A. The column was washed with approximately ten column volumes of Buffer A, followed by approximately five column volumes of 97% Buffer A,

3% Buffer B (50 mM HEPES pH 7.5, 0.2 M NaSO4, 12% glycerol, 500 mM imidazole). ThiC was eluted by a linear gradient of 3-100% Buffer B over five column volumes. Peak elution of wild-type ThiC and the variants was at approximately 45% Buffer B, or ~230 mM imidazole.

Brown fractions were pooled and desalted into Buffer A by a PD-10 Sephadex G-25 column (GE

Healthcare). The desalted protein was concentrated in a 10,000 MWCO centrifugal filter unit

(Amicon) at 2,400 x g in sealed centrifuge tubes outside of the glovebox. To allow comparison of iron loading between the ThiC variants, the proteins were not reconstituted. An aliquot of protein was removed from the glovebox and concentration was then measured by Pierce 660 assay using bovine serum albumin (BSA) as a standard, and concentration was adjusted to 500

µM. ThiC was then aliquoted into sealed serum vials for storage at -80°C. Protein concentration

39 was validated by amino acid analysis and found to be within 10 % of that measured by the Pierce

660 assay with BSA standard (Molecular Structure Facility, University of California, Davis,

CA). Wild-type ThiC and the variants were analyzed by SDS-PAGE gel and visualized by

Coomassie stain. Each preparation was >95% pure as analyzed by TotalLab gel analysis software and the dominant band was at the position expected for the MW.

Synthesis of AIR. AIR was synthesized using a modification of a previous protocol (28). In a typical synthesis, 36 mg 5-amino-4-imidazolecarboxamide ribotide (AICAR) (Sigma) was dissolved in 5 mL 4N LiOH and saponified to 5-amino-4-imidazolecarboxylic acid (CAIR) by heating in a 100-mL round bottom flask attached to a reflux condenser in a paraffin oil bath maintained at 120°C for 4 h. After removing from heat, condensate was collected with 10 mL ddH2O and the pH of the mixture was adjusted to 7.0 with 1 N acetic acid. The neutralized mixture was frozen in liquid nitrogen in two 50-mL conical tubes and lyophilized to a white powder. The powder was dissolved in 25 mL 100% ethanol and vortexed for 1 min. After 5 min, the insoluble CAIR was pelleted by centrifuging at 48,000 x g for 5 min and the supernatant, was discarded. This wash step was repeated four times before the final pellet was resuspended in 10 mL ethanol, aliquoted into eppi tubes and dried by centrifugation under vacuum. Powdered

CAIR was stored at -20°C until use.

AIR was synthesized enzymatically from CAIR by Treponema denticola AIR carboxylase (TdPurE; EC:5.4.99.18). TdPurE was expressed from pJK376, and was purified as previously described (41). The TdPurE (7.0 mg/mL, 0.4 mM) was frozen in beads in liquid nitrogen prior to storage at -80°C. To synthesize AIR, an aliquot of CAIR was dissolved in 50

µL Reaction Buffer (50 mM TAPS pH 8.0) and incubated with 20 nM TdPurE for 30 min at

30°C. The reaction was stopped by heat treatment (80°C, 3 min) and protein was pelleted by

40 centrifugation at 21,100 x g for 1 min. The AIR-containing supernatant was retained, and used in

ThiC assays within 12 h.

ThiC assays. Fpr, FldA, methylthioadenosine nucleosidase (MTAN) and AIR were degassed with nitrogen for 10 min in sealed eppendorf tubes prior to entering the glovebox. SAMe

(NatureMade) was used as the AdoMet source because it is approximately 88% biologically active S,S-AdoMet, a significant improvement over the AdoMet available from Sigma which is approximately 43% pure and contaminated by inhibiting AdoMet-related biomolecules (19).

SAMe was crushed to form a powder, then moved into the glovebox and resuspended in anaerobic Reaction Buffer (50 mM TAPS buffer pH 8.0) and passed through a Spin-X filter.

Aliquots of AIR and AdoMet were removed from the glovebox and their concentration was determined using a Nanodrop spectrophotometer (Thermo Scientific) using the extinction

-1 -1 -1 -1 coefficients ε260=1570 M cm and ε259=15400 M cm , respectively (6, 29). HPLC analysis determined SAMe was 88% pure AdoMet and the concentration in the assay was calculated accordingly.

In a typical reaction, ThiC (50 µM), MTAN (0.8 nM), Fpr (10 µM), FldA (20 µM),

NADPH (1 mM), were incubated for 10 min at room temperature before adding AdoMet (25

µM-250 µM), and AIR (25 µM-250 µM) to a final volume of 50 µL. Fpr, FldA and NADPH were provided to reduce the ThiC [Fe-S] cluster. MTAN was provided to remove

5’deoxyadenosine from the reaction mix, which has been shown to inhibit other Radical SAM enzymes (7, 19). The reactions were incubated at 37°C in the anaerobic chamber for the specified time, stopped by heat treatment (85°C for 3 min) and frozen at -20°C before analysis if necessary.

Quantification of HMP by HPLC. The heat-denatured protein was pelleted from the reaction mix by centrifugation at 21,100 x g for 1 min and the supernatant was moved to a new eppi tube.

In our hands the ThiC reaction often produces a mixture of phosphorylated and non- phosphorylated HMP; in order to convert all product to HMP, rAPid alkaline phosphatase

(Roche) was added to the supernatant (1 µL alkaline phosphatase/50 µL assay mixture) and incubated for 15 min at 37°C. The mixture was then heat treated (3 min at 80°C) to inactivate the alkaline phosphatase. The denatured enzyme was pelleted by centrifugation at 21,100 x g for 1 min and the supernatant was transferred to a HPLC autosampler vial (Macherey-Nagel). 5 µL injections of reaction mixtures were separated by RP-HPLC with a LC-20AT delivery system

(Shimadzu) equipped with a 250 x 4.6 mm Luna C18 (2), 5 micron chromatography resolution column (Phenomenex). The column was equilibrated with 90% A (20 mM potassium phosphate buffer, pH 7.0), and 10% B (methanol). The separation used a flow rate of 1 mL min-1, with 90%

A, 10% B, for 10 min followed by a linear gradient to 50% B over 20 min. Components eluted from the column were monitored with a photodiode array detector (wavelengths 190-350 nm), with data extracted 235 nm. The HMP peak was identified by a) biological activity of collected fractions; b) a UV spectrum matching that of authentic HMP; c) co-elution with authentic HMP; d) mass spectrometry. For mass spectrometry, the University of Wisconsin Biotechnology Center

Mass Spectrometry/Proteomics Facility performed standard MS in positive mode. HMP was quantified by comparison to a standard curve of authentic HMP (provided by JoAnn Stubbe,

Massachusetts Institute of Technology), which was determined using the extinction coefficient

-1 -1 ε268=4266 M cm (Parker and Stubbe, personal communication). Data are reported as average and standard deviation of duplicate reactions.

2.4 RESULTS AND DISCUSSION

Eighteen mutant alleles of thiC were isolated, sequenced and reconstructed in four different genetic backgrounds (Table 2.1). Eleven of the eighteen alleles of thiC caused a requirement for HMP or thiamine under all conditions tested. Three of these variants had substitutions in residues in the predicted active site (10): ThiCG355D; ThiCR397H; and ThiCH501Y

(Fig. 2.2). The isolation of these alleles in an unbiased screen for loss of function supported the assignment of these residues to the active site by the modeled ThiC structure (10). The other eight null variants had substitutions that were near the active site in the tertiary structure. Four of these variants, ThiCG355D, ThiCG472D, ThiCG481S and ThiCG486D were substitutions of glycine residues in α-helices, and the disruption of helical structure could explain their enzymatic defect.

The remaining seven alleles of thiC supported thiamine synthesis in vivo under certain conditions and were characterized further to understand the impact of the cellular environment on their activity. Specifically, the variants were characterized for the ability to support thiamine synthesis in genetic backgrounds and with growth conditions that had predictable consequences for the metabolic network.

Purine pathway flux impacts activity of some ThiC variants. In an otherwise wild-type genetic background, three thiC alleles allowed growth at or near wild-type levels in minimal medium, but the strains were unable to grow when adenine was present in the medium, unless thiamine was added (Table 2.2). Based on this phenotype, these variants, ThiCE281K, ThiCV267M and ThiCG92D, were designated as “adenine sensitive”. Because purine biosynthesis is regulated at both transcriptional and post-translational levels (37, 44), addition of exogenous adenine is expected to decrease flux through the pathway and therefore lower the AIR concentration in the

- 50

- 100

- 150

- 200

- 250

- 300

- 350

- 400

- 450

- 500

- 550

- 600

- 631 FIGURE 2.2 Primary sequence of ThiC with position of variants. Residues in the predicted active site, based on C. crescentus ThiC (10), were identified in S. enterica ThiC by using Clustal 2.1 sequence alignment and are outlined with a dashed line. Residues that are substituted in variants identified in this study are highlighted based on the phenotype their presence causes in vivo: null phenotype (black); conditional thiamine auxotrophs (gray). Asterix above the residues indicate that their null phenotype was characterized by site-directed mutagenesis in a previous study (13).

TABLE 2.2 Growth reflects thiamine synthesis allowed by thiC alleles.

Growth ratea (h-1) purE purE purR Alleleb Protein Ade None purR Ade Ade Ade - WT 0.58±0.01 0.68±0.06 0.53±0.01 0.50±0.04 0.64±0.04 thiC1128 ThiCE281K 0.07±0.04 0.62±0.08 0.21±0.01 0.43±0.02 0.56±0.03 thiC1129 ThiCV267M 0.07±0.03 0.54±0.06 0.20±0.07 0.49±0.01 0.16±0.03 thiC1147 ThiCG273N 0.02±0.03 0.01±0.01 0.02±0.00 0.08±0.01 0.02±0.01 thiC1158 ThiCA527T 0.01±0.01 0.02±0.01 0.13±0.02 0.39±0.03 0.02±0.02 D509G thiC1159 ThiC 0.05±0.04 0.01±0.01 0.09±0.03 0.24±0.01 0.01±0.01 thiC1161 ThiCP498L 0.01±0.00 0.03±0.01 0.17±0.02 0.30±0.07 0.01±0.01

- WTc 0.26±0.01 0.31±0.02 0.35±0.04 0.33±0.02 0.45±0.02 thiC1146 ThiCG92D,c 0.00±0.01 0.24±0.02 0.26±0.02 0.24±0.01 0.13±0.01

a Growth rate is reported as µ (µ=ln(X/Xo)/T ) where X and X0 are the A650 value during exponential growth (typically between A650 0.2 and 0.7) and T is time in h. Under all conditions the addition of thiamine resulted in a growth rate of 0.42-0.87 h-1 (data not shown). Data shown are the average and standard deviation of three independent cultures. Strains were grown at 37°C in minimal medium with no supplement or with adenine (Ade; 0.4 mM), as indicated. All data from each condition were collected on the same day. b Each allele was present in a strain with the indicated additional genotype. c Strains were grown at 30°C.

45 cell (30). Based on this assumption, these data suggested that the adenine-sensitive ThiC variants required more AIR than the wild-type protein to function in vivo.

While exogenous adenine is expected to reduce AIR levels in vivo, it potentially has additional effects on the metabolic network. To test the hypothesis that the adenine-sensitive variants were in fact sensitive to AIR levels, growth was tested in genetic backgrounds that retained high levels of AIR despite the presence of adenine. PurK (5-(carboxyamino) imidazole ribonucleotide synthase, EC:6.3.4.18) and PurE (5-(carboxyamino) imidazole ribonucleotide mutase, EC:5.4.99.18) catalyze the enzymatic steps in the purine biosynthetic pathway that consume AIR. The purE3043 allele, is expected to result in the accumulation of AIR in the presence of exogenous adenine (14, 30). The purE3043 allele restored thiamine synthesis in strains with ThiCE281K and ThiCV267M variants, as judged by the growth rates of ~0.20 h-1 compared to ~0.07 h-1 in adenine medium (Table 2.2). ThiCG92D is a temperature sensitive variant, and at 30° in the presence of purE3043, the growth rate in adenine was fully restored to that in minimal medium (~0.24 h-1). Interrupting the gene encoding purine repressor (PurR) further increased the growth rate of these strains in adenine medium, consistent with a need of the variants for higher AIR accumulation (30, 38). Finally, the ThiCD509G variant supported growth only in the purE purR genetic background, suggesting it had the highest requirement for

AIR among the variants. We interpreted these results to mean the adenine-sensitive variants required increased levels of AIR to function in vivo. Taken together, these data suggested the

ThiCE281K, ThiCV267M, ThiCD509G and ThiCG92D variants would require a higher concentration of

AIR to function in vitro compared to the wild-type enzyme.

Exogenous methionine or increased expression of AdoMet synthetase restores function to some ThiC variants. ThiC requires AdoMet as a co-substrate for radical catalysis (10, 27), so

A 1 650 nm OD

0.1

01020 Time (h) B 1 650 nm OD

0.1

01020 Time (h)

FIGURE 2.3 Either methionine or induction of S-adenosylmethionine biosynthesis allowed growth of strains containing ThiCG273N or ThiCP498L. A) Growth of the strain containing ThiCG273N (thiC1147, DM13321). B) Growth of the strain containing ThiCP498L (thiC1161, DM13224). Strains were grown in minimal medium and arabinose (0.2%) with no additions (squares), addition of methionine (triangles), or addition of thiamine (circles). Strains containing an empty vector are depicted with open symbols, while those containing pMETK2 are depicted with filled symbols. Data shown are the average and standard deviation of three cultures derived from independent colonies.

47 changes to the metabolic network that impact AdoMet were predicted to alter growth of some thiC mutant strains. Because addition of exogenous methionine to the medium has been shown to elevate AdoMet levels in yeasts (21), the thiC mutant strains were screened for growth on minimal medium with exogenous methionine. The growth of two strains, encoding variants

ThiCG273N and ThiCP498L, was greatly improved by the addition of methionine. Significantly, growth of these same strains was improved by the presence of the plasmid pMETK2, which encodes MetK (AdoMet synthetase, EC:2.1.5.6) (Fig. 2.3). The presence of pMETK2 increased

AdoMet in the cell by 5-fold (Wang and Frey, personal communication). Plasmid pMETK2 did not improve growth of a strain containing the null ThiCH501Y variant, supporting the conclusion that AdoMet, and not thiamine, was increased by the presence of this plasmid (data not shown).

These results suggested that addition of exogenous methionine increases AdoMet levels in S. enterica, as it was previously shown to do in yeast (21). Further, these in vivo phenotypes suggested the ThiCG273N and ThiCP498L variants would have a higher requirement for AdoMet than wild-type ThiC when assayed in vitro.

Exogenous pantothenate stimulates the growth rate of ThiCA527T. Conversion of AIR to

HMP-P in vivo is sensitive to the CoA levels in the cell (1). CoA has no demonstrable effect on

ThiC activity in vitro; therefore, we predicted that the CoA effect was mediated through another reaction component, i.e. AIR, AdoMet or [Fe-S] cluster. Each thiC mutant strain was screened for growth stimulation by pantothenate because exogenous pantothenate elevates endogenous

CoA levels (20). Growth of a single strain, DM13294, which carried both purE3043 and thiC1158, was stimulated by exogenous pantothenate (Fig. 2.4). This result suggested that the

ThiCA527T variant responded to the reaction component affected in vivo by CoA levels. The

1 650 nm OD

0.1

051015 Time (h)

FIGURE 2.4 The growth rate of a strain containing ThiCA527T in a purE mutant background is increased by exogenous pantothenate. Strain DM13294 (thiC1158 purE3043) was grown in NCE glucose medium supplemented with adenine (0.4 mM) and no further additions (squares), addition of pantothenate (diamonds) and addition of thiamine (circles). Data shown are the average and standard deviation of three independent cultures.

49 finding that pantothenate stimulated growth only when the purE3043 mutation was present implicated AIR concentration in the effects of CoA on ThiC activity in vivo.

ThiC variants are generally defective in activity in vitro. The differing phenotypic behaviors of the thiC mutant strains suggested that the biochemical properties of the ThiC variants would be distinguishable from each other and from those of the wild-type protein. Due to limitations of

HMP detection at low enzyme concentrations and the inability to achieve enzyme turnover, which is common for SAM radical proteins (35), Michaelis-Menten kinetic models could not be used to analyze the assay results. Wild-type and four variants of ThiC (V267M, G273N, A527T,

D509G) were purified and the activity was determined in assays using 50 µM protein, 250 µM

AIR and 250 µM AdoMet over 1 h. The data in Table 2.3 showed that with these reaction conditions, each of the variants was significantly compromised in activity. Iron content of the variants was similar to wild-type levels (data not shown); however, a conformational difference between the [Fe-S] cluster of the wild-type and mutant proteins could explain the activity difference in vivo. There was no condition where the growth rates (Table 2.2) of the strains expressing the respective thiC alleles were proportional to the ThiC variant activity in vitro.

Similarly, independent titrations of AIR and AdoMet concentrations with purified variant protein failed to provide insights about how the activity correlated with in vivo results. The activity of variants ThiCG273N and ThiCD509G was not significantly different over the range of

AIR or AdoMet used (data not shown). This result was unexpected for ThiCG273N, which allowed wild-type levels of in vivo growth when internal AdoMet levels were elevated by either methionine or pMETK (Fig. 2.3). The constant weak activity of ThiCD509G was not unexpected, given the complex nutritional environment needed to generate wild-type growth with this variant. In contrast, the activity of both ThiCV267M and ThiCA527T were affected by the

TABLE 2.3 ThiC variants have reduced activity.a

ThiC variant HMP produced (nmoles) WT 2.59 ± 0.31 ThiCV267M 0.73 ± 0.02 ThiCG273N 0.38 ± 0.05 ThiCD509G 0.43 ± 0.06 ThiCA527T 1.44 ± 0.30

a Each reaction contained the relevant ThiC protein at 50 µM, and both AdoMet and AIR were at 250 µM. Reactions were incubated for 1 h at 37°C, and heat quenched at 85°C for 3 min; detailed reaction conditions are described in Materials and Methods. Data for wild-type protein is the average and standard deviation of eight replicates. Data for each variant protein is the average and standard deviation of four replicates.

51 concentration of AIR and AdoMet (Fig. 2.5). In the case of ThiCV267M, the activity did not require an elevated level of AIR as predicted from the in vivo growth. The variants’ lower specific activity could be caused by thermal instability in the 1-h incubation; however, instability would not explain the lack of expected response to substrate concentration.

ThiCA527T had wild-type activity with low concentrations of either AIR or AdoMet, yet the best in vivo growth supported by this variant required conditions thought to increase AIR.

Exogenous pantothenate had a positive impact on the growth of strains carrying this variant, which suggested that the biochemical characteristics of the ThiCA527T variant would reveal the role of CoA. However, the ThiCA527T variant’s response to substrate concentrations and its iron content were similar to those of wild-type ThiC; thus, no particular reaction component could be implicated in the CoA effect. Further studies into the role of CoA in the ThiC reaction will screen specifically for variants that require higher CoA levels and may provide greater insight into this metabolic connection.

Conclusions. The differing in vivo behavior of thiC mutant strains indicated that the variant proteins had different capacities and/or requirements for HMP synthesis. Based on knowledge of the relevant pathways and metabolites, simple predictions about characteristics of the variant proteins were made. The finding that the variants failed to act as predicted in vitro suggested that: (a) simplistic explanations or the metabolic changes caused by mutations and/or nutrients were incorrect; or (b) the in vitro reconstitution of ThiC activity was not adequately representing in vivo conditions. These two scenarios are not mutually exclusive and we expect that the lack of correlation between in vivo growth behavior and in vitro activity reflected a combination of the two scenarios. For example, the metabolic network may be providing unanticipated metabolites and/or proteins that are crucial for optimal ThiC activity in vivo but have not yet been identified,

A 60

M) 40 μ

[HMP] ( 20

0 0255075100 [AIR] (μM)

B 60

[HMP] (µM) 20

0 0255075100 [AdoMet] (μM)

FIGURE 2.5 ThiC variants are compromised for HMP formation regardless of substrate concentration. Total HMP formed in 1 hr by 50 µM ThiC proteins with varying concentrations of substrate is shown. Data are shown for ThiC (triangles), ThiCV267M (squares) and ThiCA527T (circles). A) Reactions contained 250 µM SAM and the indicated concentration of AIR. B) Reactions contained 250 µM AIR and the indicated concentration of SAM. Data points are the average of two replicates with standard deviation.

53 and are therefore not present in the in vitro assay. Regardless of the specific explanation for the lack of correlation between in vivo and in vitro results, this study highlights the need for caution when extrapolating between in vitro and in vivo conditions. When conditions that allow ThiC to turnover in vitro have been identified, refinement of the assay can strive to detect the expected correlation between in vivo behavior and in vitro properties.

2.5 ACKNOWLEDGEMENTS

We would like to thank Mackenzie Parker and JoAnne Stubbe for providing authentic

HMP and HMP-P, their molar extinction coefficient, and guidance for HPLC separation of the

ThiC reaction products. We also acknowledge the helpful discussion about ThiC enzymology with George Reed. We thank Jannell Bazurto for the purified MTAN. We would like to thank

Nicole Buan and Jorge Escalante for plasmid pMETK2, Dennis Dean for plasmid pDB1282, and

Joseph Kappock for plasmid pJK376 and an aliquot of purified TdPurE. The National Institutes of Health grant GM47296 and the 21st Century Scientists Scholars program from the J. M.

McDonnell Foundation to DMD supported this work. LDP was supported by NSF through

Graduate Research Fellowship grant DGE-0718123.

Michael J. Dougherty isolated, sequenced and initially characterized the thiC mutant alleles. I performed all of the remaining experiments described in this chapter.

2.6 REFERENCES

1. Allen, S., J. L. Zilles, and D. M. Downs. 2002. Metabolic flux in both the purine mononucleotide and histidine biosynthetic pathways can influence synthesis of the hydroxymethyl pyrimidine moiety of thiamine in Salmonella enterica. J. Bacteriol. 184:6130-6137.

2. Backstrom, A. D., R. A. S. Mcmordie, and T. P. Begley. 1995. Biosynthesis of thiamin-I: the function of the thiE gene-product. J. Am. Chem. Soc. 117:2351-2352.

3. Balch, W. E., G. E. Fox, L. J. Magrum, C. R. Woese, and R. S. Wolfe. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260-296.

4. Bazurto, J. V., and D. M. Downs. 2011. Plasticity in the purine-thiamine metabolic network of Salmonella. Genetics 187:623-631.

5. Bennett, B. D., E. H. Kimball, M. Gao, R. Osterhout, S. J. Van Dien, and J. D. Rabinowitz. 2009. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat. Chem. Biol. 5:593-599.

6. Bock, R. M., N. S. Ling, S. A. Morell, and S. H. Lipton. 1956. Ultraviolet absorption spectra of adenosine-5'-triphosphate and related 5'-ribonucleotides. Arch. Biochem. Biophys. 62:253-264.

7. Challand, M. R., T. Ziegert, P. Douglas, R. J. Wood, M. Kriek, N. M. Shaw, and P. L. Roach. 2009. Product inhibition in the radical S-adenosylmethionine family. FEBS Lett. 583:1358-1362.

8. Chan, R. K., D. Botstein, T. Watanabe, and Y. Ogata. 1972. Specialized transduction of tetracycline resistance by phage P22 in Salmonella typhimurium. II. Properties of a high- frequency-transducing lysate. Virology 50:883-898.

9. Chatterjee, A., A. B. Hazra, S. Abdelwahed, D. G. Hilmey, and T. P. Begley. 2010. A "radical dance" in thiamin biosynthesis: mechanistic analysis of the bacterial hydroxymethylpyrimidine phosphate synthase. Angew. Chem. Int. Ed. Eng. 49:8653- 8656.

10. Chatterjee, A., Y. Li, Y. Zhang, T. L. Grove, M. Lee, C. Krebs, S. J. Booker, T. P. Begley, and S. E. Ealick. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

11. Cicchillo, R. M., K. H. Lee, C. Baleanu-Gogonea, N. M. Nesbitt, C. Krebs, and S. J. Booker. 2004. Escherichia coli lipoyl synthase binds two distinct [4Fe-4S] clusters per polypeptide. Biochemistry 43:11770-11781.

12. Davis, R. W., D. Botstein, and J. R. Roth. 1980. Advanced bacterial genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

13. Dougherty, M. J., and D. M. Downs. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

14. Dougherty, M. J., and D. M. Downs. 2004. A mutant allele of rpoD results in increased conversion of aminoimidazole ribotide to hydroxymethyl pyrimidine in Salmonella enterica. J. Bacteriol. 186:4034-4037.

15. Enos-Berlage, J. L., M. J. Langendorf, and D. M. Downs. 1998. Complex metabolic phenotypes caused by a mutation in yjgF, encoding a member of the highly conserved YER057c/YjgF family of proteins. J. Bacteriol. 180:6519-6528.

16. Estramareix, B., and S. David. 1990. Conversion of 5-aminoimidazole ribotide to the pyrimidine of thiamin in enterobacteria: study of the pathway with specifically labeled samples of riboside. Biochim. Biophys. Acta. 1035:154-160.

17. Estramareix, B., and M. Lesieur. 1969. [Biosynthesis of the pyrimidine portion of thiamine: source of carbons 2 and 4 in Salmonella typhimurium]. Biochim. Biophys. Acta. 192:375-377.

18. Estramareix, B., and M. Therisod. 1984. Biosynthesis of thiamine - 5-aminoimidazole ribotide as the precursor of all the carbon-atoms of the pyrimidine moiety. J. Am. Chem. Soc. 106:3857-3860.

19. Farrar, C. E., K. K. Siu, P. L. Howell, and J. T. Jarrett. 2010. Biotin synthase exhibits burst kinetics and multiple turnovers in the absence of inhibition by products and product-related biomolecules. Biochemistry 49:9985-9996.

20. Frodyma, M., A. Rubio, and D. M. Downs. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:236-240.

21. Holcomb, E. R., and S. K. Shapiro. 1975. Assay and regulation of S-adenosylmethionine synthetase in Saccharomyces cerevisiae and Candida utilis. J. Bacteriol. 121:267-271.

22. Hong, J., S. Sun, T. Derrick, C. Larive, K. B. Schowen, and R. L. Schowen. 1998. Transition-state theoretical interpretation of the catalytic power of pyruvate decarboxylases: the roles of static and dynamical considerations. Biochim. Biophys. Acta. 1385:187-200.

23. Hong, J. S., and B. N. Ames. 1971. Localized mutagenesis of any specific small region of the bacterial chromosome. Proc. Natl. Acad. Sci. U. S. A. 68:3158-3162.

24. Koenigsknecht, M. J., L. A. Fenlon, and D. M. Downs. 2010. Phosphoribosylpyrophosphate synthetase (PrsA) variants alter cellular pools of ribose 5- phosphate and influence thiamine synthesis in Salmonella enterica. Microbiology 156:950-959.

25. LaPorte, D. C., K. Walsh, and D. E. Koshland, Jr. 1984. The branch point effect. Ultrasensitivity and subsensitivity to metabolic control. J. Biol. Chem. 259:14068-14075.

26. Lawhorn, B. G., R. A. Mehl, and T. P. Begley. 2004. Biosynthesis of the thiamin pyrimidine: the reconstitution of a remarkable rearrangement reaction. Org. Biomol. Chem. 2:2538-2546.

27. Martinez-Gomez, N. C., and D. M. Downs. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

28. Mehl, R. A., and T. P. Begley. 2002. Synthesis of P-32-labeled intermediates on the purine biosynthetic pathway. J. Labelled Comp. Radiopharm. 45:1097-1102.

29. Meyer, E., N. J. Leonard, B. Bhat, J. Stubbe, and J. M. Smith. 1992. Purification and characterization of the purE, purK, and purC gene-products - identification of a previously unrecognized energy requirement in the purine biosynthetic-pathway. Biochemistry 31:5022-5032.

30. Petersen, L., J. Enos-Berlage, and D. M. Downs. 1996. Genetic analysis of metabolic crosstalk and its impact on thiamine synthesis in Salmonella typhimurium. Genetics 143:37-44.

31. Petersen, L. A., and D. M. Downs. 1997. Identification and characterization of an operon in Salmonella typhimurium involved in thiamine biosynthesis. J. Bacteriol. 179:4894- 4900.

32. Ramos, I., and D. M. Downs. 2003. Anthranilate synthase can generate sufficient phosphoribosyl amine for thiamine synthesis in Salmonella enterica. J. Bacteriol. 185:5125-5132.

33. Ramos, I., E. I. Vivas, and D. M. Downs. 2008. Mutations in the tryptophan operon allow PurF-independent thiamine synthesis by altering flux in vivo. J. Bacteriol. 190:815-822.

34. Raschke, M., L. Burkle, N. Muller, A. Nunes-Nesi, A. R. Fernie, D. Arigoni, N. Amrhein, and T. B. Fitzpatrick. 2007. Vitamin B1 biosynthesis in plants requires the essential iron sulfur cluster protein, THIC. Proc. Natl. Acad. Sci. U. S. A. 104:19637- 19642.

35. Roach, P. L. 2011. Radicals from S-adenosylmethionine and their application to biosynthesis. Curr. Opin. Chem. Biol. 15:267-275.

36. Roberts, G. P. 1978. Isolation and Characterization of Informational Suppressors in Salmonella typhimurium. University of California, Berkeley, Berkeley, CA.

37. Rolfes, R. J., and H. Zalkin. 1988. Escherichia coli gene purR encoding a repressor protein for purine nucleotide synthesis. Cloning, nucleotide sequence, and interaction with the purF operator. J. Biol. Chem. 263:19653-19661.

38. Rolfes, R. J., and H. Zalkin. 1988. Regulation of Escherichia coli purF. Mutations that define the promoter, operator, and purine repressor gene. J. Biol. Chem. 263:19649- 19652.

39. Salati, L. M., C. J. Gross, L. M. Henderson, and D. A. Savaiano. 1984. Absorption and metabolism of adenine, adenosine-5'-monophosphate, adenosine and hypoxanthine by the isolated vascularly perfused rat small intestine. J. Nutr. 114:753-760.

40. Schmieger, H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-88.

41. Tranchimand, S., C. M. Starks, Mathews, II, S. C. Hockings, and T. J. Kappock. 2011. Treponema denticola PurE Is a bacterial AIR carboxylase. Biochemistry 50:4623-4637.

42. Way, J. C., M. A. Davis, D. Morisato, D. E. Roberts, and N. Kleckner. 1984. New Tn10 derivatives for transposon mutagenesis and for construction of lacZ operon fusions by transposition. Gene 32:369-379.

43. Webb, E., and D. Downs. 1997. Characterization of thiL, encoding thiamin- monophosphate kinase, in Salmonella typhimurium. J. Biol. Chem. 272:15702-15707.

44. Zhou, G., J. L. Smith, and H. Zalkin. 1994. Binding of purine nucleotides to two regulatory sites results in synergistic feedback inhibition of glutamine 5- phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 269:6784-6789.

CHAPTER 3

THE THIAMINE BIOSYNTHETIC ENZYME THIC CATALYZES

MULTIPLE TURNOVERS AND IS INHIBITED BY

S-ADENOSYLMETHIONINE (ADOMET) METABOLITES1

1 Palmer, L. D., D. M. Downs. 2013. Journal of Biological Chemistry. 288:!30693-30699. Reprinted here with permission of publisher.

3.1 ABSTRACT

ThiC (4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate synthase; EC 4.1.99.17) is a radical S-adenosylmethionine (AdoMet) enzyme that uses a [4Fe-4S]+ cluster to reductively cleave AdoMet to methionine and a 5′-deoxyadenosyl radical that initiates catalysis. In plants and bacteria, ThiC converts the purine intermediate 5-aminoimidazole ribotide to 4-amino-5- hydroxymethyl-2-methylpyrimidine phosphate, an intermediate of thiamine pyrophosphate

(coenzyme B1) biosynthesis. In this study, assay conditions were implemented that consistently generated 5-fold molar excess of HMP, demonstrating that ThiC undergoes multiple turnovers.

ThiC activity was improved by in situ removal of product 5′-deoxyadenosine. The activity was inhibited by AdoMet metabolites S-adenosylhomocysteine, adenosine, 5′-deoxyadenosine, S- methyl-5′-thioadenosine, methionine, and homocysteine. Neither adenosine nor S-methyl-5′- thioadenosine had been shown to inhibit radical AdoMet enzymes, suggesting that ThiC is distinct from other family members. The parameters for improved ThiC activity and turnover described here will facilitate kinetic and mechanistic analyses of ThiC.

3.2 INTRODUCTION

ThiC (HMP-P synthase, EC 4.1.99.17) is a radical S-adenosylmethionine (AdoMet) enzyme that catalyzes the intramolecular rearrangement of 5-aminoimidazole ribotide (AIR) into

HMP-P, carbon monoxide, and formate (Fig. 3.1) (1–4). HMP-P is condensed with 4-methyl-5-

β-hydroxyethylthiazole phosphate to generate thiamine phosphate, which is further phosphorylated to the biologically active cofactor thiamine pyrophosphate (reviewed in Refs. 5,

6). Metabolically, AIR is at the branch point of purine and thiamine biosynthesis and has been known for decades to be the sole source of carbon for HMP-P (7–9). ThiC binds a [4Fe-4S]2+ cluster with a CX2CX4C motif (1), a unique variation on the canonical radical AdoMet

FIGURE 3.1 ThiC rxeaction. Abbreviations: AIR, aminoimidazole ribotide; HMP-P, 4-amino- 5-hydroxymethyl-2-methylpyrimidine phosphate; CO, carbon monoxide; HCOOH, formate.

+ superfamily motif CX3CX2C (10). Once reduced, the [4Fe-4S] cluster reductively cleaves

AdoMet, producing methionine (Met) and a 5′-deoxyadenosyl radical that initiates catalysis. In vitro work by Chatterjee et al. (4) suggested that ThiC catalysis used two sequential hydrogen abstractions by the 5′-deoxyadenosyl radical, a mechanism that had not been reported.

Numerous radical AdoMet enzymes have been identified by bioinformatics analysis, and those that have been characterized carry out diverse reactions within metabolism, including nucleic acid modification and repair and synthesis of cofactors and antibiotics. Enzymes in the radical AdoMet superfamily can be divided into three classes (11–13). The first class uses

AdoMet as a catalytic cofactor and includes spore photoproduct lyase and lysine 2,3- aminomutase (14, 15). The second class is made up of glycyl radical-activating enzymes that catalyze radical formation on glycines in other enzymes. This class includes pyruvate formate- lyase activating enzyme and ribonucleotide reductase-activating enzyme (16, 17). Enzymes in the third class use AdoMet as a substrate. The majority of radical AdoMet enzymes characterized to date fall into this class, including lipoyl synthase, tyrosine lyase, and biotin synthase (BioB)

(18–20).

According to the literature, ThiC uses AdoMet as an oxidizing cosubstrate (1:1 stoichiometry) (4), making it a member of the third class described above. The activities of BioB, tyrosine lyase, and lipoyl synthase are inhibited by AdoMet cleavage products 5′- deoxyadenosine (5′-DOA) and Met (21, 22), whereas other enzymes in this class (including the maturase from Klebsiella pneumoniae AtsB and butirosin biosynthetic enzyme BtrN) are not product-inhibited (23, 24). S-methyl-5′-thioadenosine nucleosidase (MTAN, E.C. 3.2.2.9,

3.2.2.16) breaks down 5′-DOA to adenine and 5′-deoxyribose (25) and can improve activity when added to the assay mix of enzymes that are inhibited by 5′-DOA (21, 22).

This study was motivated by our interest in the complex metabolic context of the ThiC reaction in Salmonella enterica. In this organism, the conversion of AIR to HMP-P was decreased by perturbations in other metabolic processes, including the biosynthetic pathways for purines, Met, iron-sulfur clusters, and CoA (26–29). We sought to improve the in vitro assay for

ThiC activity to allow us to obtain kinetic parameters that could help us rationalize the diverse metabolic connections identified in vivo. Here, we report assay conditions for the in vitro ThiC reaction that resulted in multiple turnovers and allowed the first kinetic measurements of this enzyme activity.

3.3 EXPERIMENTAL PROCEDURES

Media and Chemicals. Difco Luria Bertani (20 g/L) medium was used for routine Escherichia coli growth. For protein overexpression, Superbroth (tryptone (32 g/liter), yeast extract (20 g/liter), and NaCl (5 g/liter) with NaOH (0.05 N)) was used. Ampicillin and kanamycin were added to the medium as needed at 150 mg/liter and 50 mg/liter, respectively. Unless noted otherwise, all chemicals were purchased from Sigma-Aldrich, St. Louis, MO.

Protein Purification and ThiC Reconstitution. Proteins flavoprotein reductase (Fpr, E.C.

1.18.1.2), flavodoxin A (FldA), and TdPurE were expressed and purified as described previously

(26, 30). TdPurE is Treponema denticola AIR carboxylase (E.C. 5.4.99.18) and was produced from pJK376 (a gift from J. Kappock). All ThiC purifications and manipulations were carried out in an anoxic glove box (Coy Laboratories, Grass Lake, MI) maintained at < 2 ppm O2. S. enterica His6-ThiC was produced from vector pET-28b(+) in a strain overexpressing Azotobacter vinelandii [Fe-S] cluster-loading genes from plasmid pDB1282 (31). ThiC was purified as described (26), except that the [4Fe-4S] cluster was reconstituted in vitro prior to freezing the

63 protein at −80 °C. After purification, ThiC concentration was determined by Pierce 660 assay

(Thermo Scientific, Rockford, IL) using BSA as the standard.

ThiC protein was reduced by adding a 50-fold excess of DTT in a vial that was then sealed and incubated on ice in the glove box overnight. A fresh stock solution of FeNH3SO4 (400 mm) was added in four aliquots to be 8-fold in excess of ThiC, and the vial was incubated at room temperature for 5 min. A fresh stock solution of Na2S (400 mm) was then added in four aliquots to reach an 8-fold excess over ThiC. Reduced ThiC was incubated for 1 h before desalting into freezing buffer (50 mm N-tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid sodium-potassium salt, 3-{[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]amino}-1- propanesulfonic acid (TAPS) (pH 8.0), 0.2 m Na2SO4, 1.6 m glycerol) by a PD-10 Sephadex G-

25 column (GE Healthcare Life Sciences, Piscataway, NJ). The desalted protein was concentrated in an Amicon 10,000 Da molecular weight cut off centrifugal filter unit (Millipore,

Billerica, MA) at 2400 × g in sealed centrifuge tubes outside of the glove box. The protein concentration after reconstitution was 0.27 ± 0.03 mm, as determined by Bradford assay using purified ThiC with the concentration determined by amino acid analysis as a standard.

Iron Content Determination. The iron content of the ThiC protein was determined by a colorimetric assay using 3-(2-pyridyl)-5,6-bis(5-sulfo-2-furyl)-1,2,4-triazine disodium salt trihydrate adapted from Kennedy et al. (32). All reagents were prepared in double-distilled water and in new glassware or plasticware to prevent iron contamination. 25 µl of ThiC sample dilutions and iron standard solutions (Sigma) were mixed with 25 µl of HCl (0.12 N) in 1.5-ml microcentrifuge tubes and shaken gently. After incubation at 80 °C for 10 min, reagents were added to each tube sequentially with vortexing after each addition: 125 µl of ammonium acetate

(0.96 m), 25 µl of ascorbic acid (0.2 m), 25 µl of sodium dodecyl sulfate (87 mm), and 25 µl of

3-(2-pyridyl)-5,6-bis(5-sulfo-2-furyl)-1,2,4-triazine disodium salt trihydrate (30 mm). The samples were then centrifuged for 5 min at 9000 × g, and the supernatant was analyzed for absorbance at 593 nm using a SpectraMax plate spectrophotometer (Molecular Devices,

Sunnyvale, CA) because 3-(2-pyridyl)-5,6-bis(5-sulfo-2-furyl)-1,2,4-triazine disodium salt trihydrate absorbs at 593 nm when complexed to Fe2+. Iron content was 4.2 ± 0.7 mol iron/mol

ThiC in the preparation used in the studies described herein.

Synthesis of CAIR, 4-Carboxyaminoimidazole Riboside, AIR, and Aminoimidazole

Carboxamide Ribotide. CAIR and AIR were synthesized as described (26, 30, 33). The molar extinction coefficients ϵ250 = 10,980 M/cm and ϵ250 = 3,270 M/cm (34) were used to determine their respective concentrations. 4-carboxyaminoimidazole riboside and 5-aminoimidazole riboside were generated from stocks of CAIR and AIR (10–15 mm) treated with rAPid alkaline phosphatase (Roche) at 37° for 15 min. The alkaline phosphatase was then heat-inactivated by incubating at 80° for 3 min and cleared by centrifugation at 21,100 × g for 1 min. The supernatant was transferred to a new tube and degassed for 10 min before transfer to the glove box.

Purification of AdoMet from a Pharmaceutical Source. Commercial sources of AdoMet have been found to be as little as 43% biologically active S,S-AdoMet (22). We found previously that, of the compounds absorbing at 259 nm, SAMe (NatureMade, Mission Hills, CA) was ∼88% S,S-

AdoMet by HPLC analysis (26). To further purify S,S-AdoMet, a SAMe pill was crushed, dissolved in double-distilled H2O, and filtered through a 0.22-µm Spin-X filter (Corning). The concentration of adenine compounds was determined using the extinction coefficient ϵ259 =

15,400 M/cm (35), and the concentration was adjusted to 100 mm in double-distilled H2O. 3-µl injections of the SAMe solution were separated by reverse phase-HPLC with a LC-20AT

65 delivery system (Shimadzu, Kyoto, Japan) equipped with a 250 × 4.6 mm Luna C18 (2), 5-µm chromatography resolution column (Phenomenex, Torrance, CA). The column was equilibrated with 90% mobile phase A (13 mm TFA) and 10% mobile phase B (methanol). The separation used a flow rate of 1 ml/min with 90% A, 10% B for 10 min, followed by a linear gradient to

50% B over 20 min. Components eluted from the column were monitored with a SPD-M20A photodiode array detector (Shimadzu, wavelengths 190–350 nm) with data extracted at 259 nm.

The 3.00- to 3.85-min fraction was collected using the FRC-10A fraction collector (Shimadzu) outfitted with a extruded polystyrene foam box filled with dry ice so the purified AdoMet was immediately frozen as it was collected in a 50-ml conical tube. The purified AdoMet was lyophilized and resuspended in double-distilled H2O sequentially three times to remove residual

TFA. The purified AdoMet powder was resuspended in double-distilled H2O (∼22 mm), and samples were frozen at −20° until use. HPLC analysis determined that the purified AdoMet was

99% pure.

ThiC Activity Assays. Fpr, FldA, MTAN, AdoMet, and AIR were degassed with nitrogen for 10 min in 1.5-ml microcentrifuge tubes sealed with rubber stoppers prior to being placed in the glove box. Concentrations of AIR and AdoMet were determined with a Nanodrop spectrophotometer (Thermo Scientific) using the extinction coefficients listed above.

All components were resuspended in anoxic reaction buffer (50 mm TAPS (pH 8.0)). Each assay included ThiC (0.55 nmol monomer, 11 µm), MTAN (as indicated, 0.1 nmol), Fpr (0.5 nmol), and FldA (1 nmol). Under these conditions, HMP production was linear with respect to ThiC concentration, and MTAN, Fpr, and FldA were not rate-limiting. Reduced NADPH (0.8 mm) was added in excess, and the reaction mix was incubated for 10 min at room temperature before adding the substrate of interest. Substrates AdoMet (25–150 µm) and AIR (25–150 µm) were

66 added to a final volume of 50 µl. The reactions were incubated at 37 °C in the anaerobic chamber for the specified time, stopped by heat treatment (65 °C for 3 min), and frozen at −20 °C if they were not analyzed immediately.

When included, inhibitors were preincubated with the ThiC reaction mixture for 10 min before the relevant substrates were added. Homocysteine, aminoimidazole carboxamide, Met, adenosine, and imidazole were brought into the glove box as powders and resuspended in anoxic reaction buffer. Adenosine was heated at 65 °C for 5 min to dissolve. All other potential inhibitors were made in reaction buffer, adjusted to pH 6–9, and degassed for 10 min prior to entering the glove box. In assays where we titrated specific inhibitors, the concentration of the inhibitor was determined after degassing using the relevant extinction coefficient. MTAN was not used in assays addressing inhibition.

HMP-P was dephosphorylated to HMP by alkaline phosphatase and quantified as described (26). In addition, samples were filtered through a 10,000- to 50,000-kDa cellulose membrane with an Amicon centrifugal filter (Millipore) to remove proteins prior to transferring the samples to autosampler vials (Macherey-Nagel, Düren, Germany).

Kinetic Data Analysis. Graphs were prepared, and data were analyzed using least squares analysis in Prism v. 6.0b (GraphPad Software Inc., La Jolla, CA). Kinetic constants are reported with the S.E. of the fit unless noted otherwise noted. For time course experiments, the data were fitted to a first-order kinetic equation, Equation 3.1, where [HMP] was the observed HMP produced (µm), [HMP]max was the predicted maximum HMP produced (µm), k was the observed first-order rate constant, and t was time in min.

!!" HMP = HMP !"# 1 − ! (3.1)

0 The initial turnover number, kcat , was determined by Equation 3.2, based on the methods of

Challand et al. (36).

! ! = ![!"#]!"# (3.2) !"# [!"#!!!"#"!$%]

To determine the kinetics of ThiC inhibition, the initial velocity (v; nmol HMP/nmol

ThiC/min) was estimated from reactions stopped after 20 min incubation at 37°C. The Km was determined from data titrating AdoMet (20-150 µM) and omitting MTAN. The data were fit to

Equation 3.3.

! = !!!"#[!] (3.3) !!![!]

Data were first diagnosed as competitive, uncompetitive or noncompetitive inhibition by their appearance when graphed as double reciprocal Lineweaver-Burk plots and fit by linear regression. The data were then analyzed according to the appropriate equation. For competitive inhibition, Equation 4 was used, where v is the velocity in nmol HMP/nmol ThiC/min; Vmax is the maximum velocity observed; KmObs is determined by the equation KmObs = Km(1 + [I]/Ki); and

[S] is the concentration of substrate provided.

! ![!] ! = !!"# (3.4) ! !!!"#[!]

For cooperative competitive inhibition by two different nonexclusive inhibitors, the data were fit to Equation 5 (37), where v is the velocity in nmol HMP/nmol ThiC/min; Vmax is the maximum velocity; [S] is the concentration of substrate; Ks is the Michaelis-Menten constant for the substrate; [I] is the concentration of one inhibitor, and Ki is its inhibition constant; and [X] is the concentration of the second inhibitor, and Kx is its inhibition constant; and α is the cooperativity factor.

[!] ! !! = [!] [!] [!] ! [!] (3.5) !!"# !! ! ! ! !! !! !! !!!!!

For uncompetitive inhibition, Equation 6 was used, where v is the velocity in nmol HMP/nmol

ThiC/min; VmaxApp is the apparent maximum velocity; KmApp is the apparent Km; and the Ki’ inhibition constant is determined by the equations VmaxApp = Vmax/(1+[I]/(Ki’)) and KmApp = Km/(1

+ [I]/(Ki’)).

!! ! ! ! = !"" (3.6) ! !!"#!"" !

3.4 RESULTS AND DISCUSSION

ThiC is a multiple-turnover enzyme. ThiC activity assays described elsewhere required high protein concentration and/or long incubations to quantify HMP-P production (1–4, 26). These conditions prevented mechanistic and kinetic analysis of ThiC. Changes were made to the assay protocol for ThiC to increase HMP production. The [4Fe-4S] cluster in ThiC was reconstituted in vitro, and pure sources of substrates AIR and AdoMet (99% pure) were used in the assay. With these modifications, ThiC produced 3.1 ± 0.1 nmol HMP/nmol ThiC monomer in 2 h, confirming that multiple turnovers were possible in vitro (Fig. 3.2). Under these conditions, steady-state turnover continued for 25 min. Data from technical duplicates were fit to the first- order kinetic equation (Equation 3.1) with a goodness of fit R2 of > 0.95. These results were then used in Equation 3.2 and yielded the following turnover number representing the mean ± S.E. of

0 −1 the constants determined by two independent experiments: kcat = 0.074 ± 0.014 min .

The production of HMP was significantly enhanced by the addition of MTAN. When 0.1 nmol MTAN was included in the reaction mix, ThiC produced 5.2 ± 0.1 nmol HMP/nmol ThiC, and steady-state turnover continued for 1 h. The kinetic analysis of these data yielded the

0 −1 0 turnover number kcat = 0.14 ± 0.03 min . This value for kcat is within the range reported for other radical AdoMet enzymes in this class (22–24, 36, 38).

AdoMet-related metabolites inhibit the ThiC in vitro reaction. The finding that MTAN increased the reaction rate by ∼2-fold suggested that ThiC was inhibited by its product 5′-DOA.

This conclusion was verified and extended by screening a number of potentially relevant metabolites for an effect on ThiC activity. Potential inhibitors tested included AdoMet-related metabolites, purines related to the substrate AIR, aminoimidazole carboxamide ribotide

(AICAR)-related metabolites and CoA metabolites. The latter two represented metabolic pathways shown to impact the AIR to HMP-P conversion in vivo (28, 29).

Under the conditions tested, we saw no inhibition by purine biosynthetic intermediates related to AIR, including imidazole and the AIR riboside. These data support the conclusion that the in vivo findings reflect indirect metabolic effects of AICAR and CoA on the ThiC reaction.

In contrast, several AdoMet-related metabolites inhibited ThiC, specifically 5′-DOA, Met, homocysteine, adenosine (Ado), S-adenosylhomocysteine (SAH), and S-methyl-5′-thioadenosine

(MTA) (Fig. 3.3). Of these metabolites, 5′-DOA, Met, homocysteine, and SAH are known inhibitors of radical AdoMet enzymes (reviewed in Ref. 13). The data also showed that 5′-DOA acted additively with either Met or homocysteine to further inhibit ThiC activity.

S-Adenosylhomocysteine inhibits ThiC competitively with AdoMet. SAH has been reported to inhibit representatives of all three classes of radical AdoMet enzymes: lysine 2,3- aminomutase, ribonucleotide reductase-activating enzyme, BioB, and the nitrogenase cofactor biosynthetic enzyme NifB (22, 39–41). The mechanism of SAH inhibition of ThiC was investigated by adding SAH at different concentrations (0, 10, 25, and 50 µm) to reaction mixtures containing several concentrations of AdoMet (25–150 µm). The Lineweaver-Burk plot

6 +MTAN

4 - MTAN

2 nmol HMP/nmol ThiC HMP/nmol nmol

0 060120 Time (min)

FIGURE 3.2 ThiC undergoes steady-state turnover. ThiC (0.55 nmol monomer) was incubated with flavoprotein reductase (0.5 nmol), flavodoxin A (1 nmol), NADPH (0.8 mM), AdoMet (100 µM), and AIR (100 µM) and MTAN as indicated (0.1 nmol) at 37°C. Each data point represents the average and standard deviation of two replicates from a single experiment. The data were fit to a first-order rate equation, and the 95% confidence intervals of the regression analysis are represented by dotted lines.

of these data showed that SAH inhibited ThiC competitively with AdoMet (Fig. 3.4). The Km of

ThiC for AdoMet was determined by fitting data to Equation 3.3 from duplicate reactions of a titration of AdoMet (20–150 µm) carried out without MTAN. The data were fit with a global R2 value of 0.89, and the Km was 17 ± 3 µm. On the basis of the diagnosis of competitive inhibition,

2 the data were fit to Equation 3.4 using the above Km with a global R value of 0.85 and generated

SAH the kinetic constant Ki = 5.6 ± 1.1 µm.

In the cell, SAH is produced by AdoMet methyltransferases and hydrolyzed by MTAN

(42). SAH is present at ∼1 µm in wild-type E. coli and 50 µm in a mutant strain without MTAN

(43). Together, these data suggest SAH could have a physiologically relevant role in regulating

ThiC activity under conditions where MTAN activity is reduced.

5’-deoxyadenosine and methionine cooperatively inhibit ThiC. 5′-DOA and Met were found to cooperatively inhibit BioB (22), and data from our inhibitor screen indicated that they also cooperatively inhibited ThiC. The reduction in activity by the addition of 5′-DOA and Met together (12% of activity with no inhibitor) was slightly greater than expected for linear combination of the inhibition caused by 5′-DOA (31%) or Met (55%) when either was the sole addition. To investigate the kinetics of this inhibition, several concentrations of 5′-DOA (0–500

µm) and Met (0–1000 µm) were added to ThiC reactions with AIR and AdoMet fixed at 100 µm

(Fig. 3.5). Dixon replots of 1/v versus [5′-DOA] or [Met] intersected, confirming that 5′-DOA and Met were not mutually exclusive (37). 5′-DOA and Met were assumed to inhibit competitively with respect to AdoMet. The least squares analysis was constrained to [S] = 100

2 µm and Km = 17 µm and the data fit Equation 3.5 with a global R value of 0.94 and yielded

5′−DOA Met Ki = 12 ± 2 µm, Ki = 82 ± 13 µm, and α = 0.4 ± 0.1.

1 * * * * * *

nmol HMP/nmol ThiC HMP/nmol nmol * * 0 Met Ado Ade HCy ATP CoA SAH MTA ADP AMP AIRs AICA CAIR CAIRs cAMP AICAR 5'-DOA 2'-DOA AICARs Imidazole Acetyl-CoA No Inhibitor 5'DOA+Met 5'DOA+HCy

FIGURE 3.3 Metabolite inhibitors of ThiC activity. ThiC (0.55 nmol monomer) was preincubated with flavoprotein reductase (0.5 nmol), flavodoxin A (1 nmol), NADPH (0.8 mM) and potential inhibitor (0.5 mM) for 10 min at room temperature. Then AdoMet (100 µM), and AIR (100 µM) were added to initiate the reactions, which were incubated at 37°C for 30 min. Data represent the average and standard deviation of two replicates. An asterisk indicates that the average is significantly different than the average with no inhibitor, as determined by an unpaired t-test (p<0.05). Abbreviations: 4-amino-5-hydroxymethyl-2-methylpyrimidine, HMP; 5’-deoxyadenosine, 5’-DOA; methionine, Met; homocysteine, HCy; 2’-deoxyadenosine, 2’- DOA; adenosine, Ado; adenine, Ade; methionine, Met; S-adenosylhomocysteine, SAH; S- methyl-5’-thioadenosine, MTA; 5-aminoimidazole riboside, AIRs; aminoimidazole carboxamide ribotide, AICAR; aminoimidazole carboxamide riboside, AICARs; aminoimidazole carboxamide, AICA; 5-Amino-4-imidazolecarboxylic acid ribotide, CAIR; 5-Amino-4- imidazolecarboxylic acid riboside, CAIRs.

[SAH] (μM) 80 50

25 min/nmolHMP)

• 40

10 20 (nmol ThiC v

1/ 0

0 0.00 0.01 0.02 0.03 0.04 1/[AdoMet] (μM-1)

FIGURE 3.4 SAH inhibits ThiC competitively with respect to AdoMet. ThiC (0.55-nmol monomer) was preincubated with flavoprotein reductase (0.5 nmol), flavodoxin A (1 nmol), NADPH (0.8 mm), AIR (100 µm), and SAH (0, 10, 25, or 50 µm) for 10 min at room temperature. Then AdoMet (25–150 µm) was added to initiate the reactions, which were incubated at 37 °C for 20 min. The data were fit to Equation 3.4 by non-linear regression, constraining Km = 17 µm.

Under normal metabolic conditions, product inhibition would be expected to be minimal. Met concentrations are estimated at 150–300 µm (43, 44), and MTAN is present to rapidly hydrolyze low levels of 5′-DOA produced. However, these constants suggest that product inhibition could be significant in in vitro assays, including those reported here. For example, after 2 h of incubation, product accumulation coupled with substrate depletion would cause ThiC to be 60% or 35% maximal activity with or without MTAN, respectively. These findings suggest that long incubation times will not allow accurate kinetic measurements of ThiC.

Adenosine displays uncompetitive inhibition with AdoMet. If adenosine bound the site occupied by the adenosine moiety of AdoMet, adenosine should also inhibit ThiC competitively with respect to AdoMet. Adenosine was added at several concentrations (0, 100, 250, and 400

µm) to reactions containing several AdoMet concentrations (25–150 µm). Unexpectedly, the data with and without adenosine resulted in parallel lines in the Lineweaver-Burk plot (Fig. 3.6A), suggesting that adenosine was uncompetitive with AdoMet and bound the ThiC-AdoMet

2 complex. To determine Vmax, the [Ado] = 0 µm data were fit to Equation 3.3 with an R value of

0.92 to yield Vmax = 0.1128 ± 0.0034 nmol HMP/nmol ThiC/min. The full dataset was fit to

Equation 3.6, constraining the Km = 17 µm and Vmax = 0.1128 nmol HMP/nmol ThiC/min. The

2 Ado data fit Equation 3.6 with a global R value of 0.91 and produced the kinetic constant Ki' = 99

± 3 µm. However, the uncertainty in the inhibition constant is likely considerably higher. We

Ado found that Ki' values of 55–140 µm were consistent with the data. Replots of the data from the reciprocal Lineweaver-Burk plot were also linear, confirming the diagnosis of uncompetitive inhibition (37). Experiments addressing adenosine inhibition with respect to AIR showed that adenosine is not competitive with AIR, which is consistent with the fact that AMP does not

1 max

V [Met] (μM) / v 0 10 25 50 100 200 0 500 0250500 [5'-DOA] (μM)

FIGURE 3.5 Cooperative inhibition by 5′-DOA and Met. ThiC (0.55-nmol monomer) was preincubated with flavoprotein reductase (0.4 nmol), flavodoxin A (1 nmol), NADPH (0.8 mm), 5′-DOA (0–500 µm), and Met (0–1000 µm) for 10 min at room temperature. Then AdoMet (100 µm) and AIR (100 µm) were added to initiate the reactions, which were incubated at 37 °C for 20 min. The data were fit to Equation 3.5 by non-linear regression, constraining Km = 17 µm and [AdoMet] = 100 µm.

76 inhibit. The data did not distinguish between uncompetitive and noncompetitive inhibition (data not shown).

The adenosine concentration in E. coli was estimated at 0.13 µm (44), suggesting that adenosine inhibition is not physiologically relevant. However, direct inhibition of ThiC may be significant under conditions of increased adenosine levels, such as with AICAR accumulation

(45) or when adenosine is present in the growth medium (46).

Conclusions. ThiC is the HMP-P synthase required for thiamine biosynthesis in bacteria and plants and is a member of the radical AdoMet superfamily of enzymes. Of numerous radical

AdoMet enzymes predicted by bioinformatic analyses, relatively few have been characterized, and fewer still have been shown to turnover catalytically in vitro (10, 38, 47). The data presented here demonstrate that when product inhibition is relieved, ThiC undergoes steady-state turnover for up to 1 h.

To our knowledge, there are no other reports of radical AdoMet enzymes inhibited by adenosine or MTA, suggesting that this may be a unique property of ThiC. Although not many enzymes have been tested, BioB was not inhibited by adenosine or MTA (22), and MTA was reported to have no effect on lysine 2,3-aminomutase activity (39). Thus, ThiC has a distinct inhibitor profile in addition to its variant cysteine motif and proposed novel catalytic mechanism.

The characterization of ThiC activity presented here, in particular achieving catalytic turnover in vitro, will contribute to future mechanistic studies of ThiC and further our understanding of the radical AdoMet enzyme superfamily.

3.5 ACKNOWLEDGEMENTS

We thank Jorge C. Escalante-Semerena for critical reading of the manuscript and George

H. Reed and Michael G. Thomas for helpful discussions. We also thank T. Joseph Kappock for

[Ado] (μM)

400

250 40

100 min/nmolHMP) •

20 0 (nmol ThiC v 1/

0 0.00 0.01 0.02 0.03 0.04 1/[AdoMet] (μM-1)

FIGURE 3.6 Adenosine is uncompetitive with AdoMet inhibiting ThiC. A. ThiC (0.55 nmol monomer) was pre-incubated with flavoprotein reductase (0.5 nmol), flavodoxin A (1 nmol), NADPH (0.8 mM), AIR (100 µM) and adenosine (0, 100, 250 or 400 µM) for 10 min at room temperature. Then AdoMet (25-150 µM) was added to initiate the reactions, which were incubated at 37°C for 20 min. The data were fit to Equation 6 by non-linear regression. Abbreviations: adenosine, Ado.

78 pJK376 expressing TdPurE, Jannell V. Bazurto for providing MTAN, and Mackenzie J. Parker and JoAnne Stubbe for authentic HMP used for quantification.This work was supported, in whole or in part, by National Institutes of Health Grant GM47296 (to D. M. D.). This work was also supported by National Science Foundation Graduate Research Fellowship grant DGE-

0718123 (to L. D. P.).

3.6 REFERENCES

1. Martinez-Gomez NC, Downs DM. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

2. Martinez-Gomez NC, Poyner RR, Mansoorabadi SO, Reed GH, Downs DM. 2009. Reaction of AdoMet with ThiC generates a backbone free radical. Biochemistry 48:217- 219.

3. Chatterjee A, Li Y, Zhang Y, Grove TL, Lee M, Krebs C, Booker SJ, Begley TP, Ealick SE. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

4. Chatterjee A, Hazra AB, Abdelwahed S, Hilmey DG, Begley TP. 2010. A "radical dance" in thiamin biosynthesis: mechanistic analysis of the bacterial hydroxymethylpyrimidine phosphate synthase. Angew. Chem. Int. Ed. Eng. 49:8653- 8656.

5. Begley TP, Downs DM, Ealick SE, McLafferty FW, Van Loon AP, Taylor S, Campobasso N, Chiu HJ, Kinsland C, Reddick JJ, Xi J. 1999. Thiamin biosynthesis in prokaryotes. Arch. Microbiol. 171:293-300.

6. Begley TP, Chatterjee A, Hanes JW, Hazra A, Ealick SE. 2008. Cofactor biosynthesis-- still yielding fascinating new biological chemistry. Curr. Opin. Chem. Biol. 12:118-125.

7. Estramareix B, David S. 1990. Conversion of 5-aminoimidazole ribotide to the pyrimidine of thiamin in enterobacteria: study of the pathway with specifically labeled samples of riboside. Biochim. Biophys. Acta 1035:154-160.

8. Estramareix B, Lesieur M. 1969. Biosynthesis of the pyrimidine portion of thiamine: source of carbons 2 and 4 in Salmonella typhimurium. Biochim. Biophys. Acta 192:375- 377.

9. Estramareix B, Therisod M. 1984. Biosynthesis of thiamine: 5-aminoimidazole ribotide as the precursor of all the carbon-atoms of the pyrimidine moiety. J. Am. Chem. Soc. 106:3857-3860.

10. Sofia HJ, Chen G, Hetzler BG, Reyes-Spindola JF, Miller NE. 2001. Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res. 29:1097-1106.

11. Frey PA, Booker SJ. 1999. Radical intermediates in the reaction of lysine 2,3- Aminomutase, p. 1-43. In Zard SZ (ed.), Advances in Free Radical Chemistry, vol. 2. JAI Press Inc., Stamford, CT.

12. Booker SJ. 2009. Anaerobic functionalization of unactivated C-H bonds. Curr. Opin. Chem. Biol. 13:58-73.

13. Hiscox MJ, Driesener RC, Roach PL. 2012. Enzyme catalyzed formation of radicals from S-adenosylmethionine and inhibition of enzyme activity by the cleavage products. Biochim. Biophys. Acta 1824:1165-1177.

14. Cheek J, Broderick JB. 2002. Direct H atom abstraction from spore photoproduct C-6 initiates DNA repair in the reaction catalyzed by spore photoproduct lyase: evidence for a reversibly generated adenosyl radical intermediate. J. Am. Chem. Soc. 124:2860-2861.

15. Moss ML, Frey PA. 1990. Activation of lysine 2,3-aminomutase by S- adenosylmethionine. J. Biol. Chem. 265:18112-18115.

16. Wagner AF, Frey M, Neugebauer FA, Schafer W, Knappe J. 1992. The free radical in pyruvate formate-lyase is located on glycine-734. Proc. Natl. Acad. Sci. U.S.A. 89:996- 1000.

17. Tamarit J, Mulliez E, Meier C, Trautwein A, Fontecave M. 1999. The anaerobic ribonucleotide reductase from Escherichia coli. The small protein is an activating enzyme containing a [4Fe-4S](2+) center. J. Biol. Chem. 274:31291-31296.

18. Miller JR, Busby RW, Jordan SW, Cheek J, Henshaw TF, Ashley GW, Broderick JB, Cronan JE, Jr., Marletta MA. 2000. Escherichia coli LipA is a lipoyl synthase: in vitro biosynthesis of lipoylated pyruvate dehydrogenase complex from octanoyl-acyl carrier protein. Biochemistry 39:15166-15178.

19. Kriek M, Martins F, Leonardi R, Fairhurst SA, Lowe DJ, Roach PL. 2007. Thiazole synthase from Escherichia coli: an investigation of the substrates and purified proteins required for activity in vitro. J. Biol. Chem. 282:17413-17423.

20. Ollagnier-de-Choudens S, Mulliez E, Fontecave M. 2002. The PLP-dependent biotin synthase from Escherichia coli: mechanistic studies. FEBS Lett. 532:465-468.

21. Challand MR, Ziegert T, Douglas P, Wood RJ, Kriek M, Shaw NM, Roach PL. 2009. Product inhibition in the radical S-adenosylmethionine family. FEBS Lett. 583:1358- 1362.

22. Farrar CE, Siu KK, Howell PL, Jarrett JT. 2010. Biotin synthase exhibits burst kinetics and multiple turnovers in the absence of inhibition by products and product-related biomolecules. Biochemistry 49:9985-9996.

23. Grove TL, Lee KH, St Clair J, Krebs C, Booker SJ. 2008. In vitro characterization of AtsB, a radical SAM formylglycine-generating enzyme that contains three [4Fe-4S] clusters. Biochemistry 47:7523-7538.

24. Yokoyama K, Numakura M, Kudo F, Ohmori D, Eguchi T. 2007. Characterization and mechanistic study of a radical SAM dehydrogenase in the biosynthesis of butirosin. J. Am. Chem. Soc. 129:15147-15155.

25. Choi-Rhee E, Cronan JE. 2005. A nucleosidase required for in vivo function of the S- adenosyl-L-methionine radical enzyme, biotin synthase. Chem. Biol. 12:589-593.

26. Palmer LD, Dougherty MJ, Downs DM. 2012. Analysis of ThiC variants in the context of the metabolic network of Salmonella enterica. J. Bacteriol. 194:6088-6095.

27. Dougherty MJ, Downs DM. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

28. Allen S, Zilles JL, Downs DM. 2002. Metabolic flux in both the purine mononucleotide and histidine biosynthetic pathways can influence synthesis of the hydroxymethyl pyrimidine moiety of thiamine in Salmonella enterica. J. Bacteriol. 184:6130-6137.

29. Frodyma M, Rubio A, Downs DM. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar typhimurium. J. Bacteriol. 182:236-240.

30. Tranchimand S, Starks CM, Mathews, II, Hockings SC, Kappock TJ. 2011. Treponema denticola PurE is a bacterial AIR carboxylase. Biochemistry 50:4623-4637.

31. Cicchillo RM, Lee KH, Baleanu-Gogonea C, Nesbitt NM, Krebs C, Booker SJ. 2004. Escherichia coli lipoyl synthase binds two distinct [4Fe-4S] clusters per polypeptide. Biochemistry 43:11770-11781.

32. Kennedy MC, Kent TA, Emptage M, Merkle H, Beinert H, Munck E. 1984. Evidence for the formation of a linear [3Fe-4S] cluster in partially unfolded aconitase. J. Biol. Chem. 259:14463-14471.

33. Mehl RA, Begley TP. 2002. Synthesis of P-32-labeled intermediates on the purine biosynthetic pathway. J. Labelled Comp. Radiopharm. 45:1097-1102.

34. Meyer E, Leonard NJ, Bhat B, Stubbe J, Smith JM. 1992. Purification and characterization of the purE, purK, and purC gene-products - identification of a previously unrecognized energy requirement in the purine biosynthetic-pathway. Biochemistry 31:5022-5032.

35. Bock RM, Ling NS, Morell SA, Lipton SH. 1956. Ultraviolet absorption spectra of adenosine-5'-triphosphate and related 5'-ribonucleotides. Arch. Biochem. Biophys. 62:253-264.

36. Challand MR, Martins FT, Roach PL. 2010. Catalytic activity of the anaerobic tyrosine lyase required for thiamine biosynthesis in Escherichia coli. J. Biol. Chem. 285:5240- 5248.

37. Segel IH. 1975. Enzyme Kinetics. John Wiley & Sons, Inc., New York.

38. Roach PL. 2011. Radicals from S-adenosylmethionine and their application to biosynthesis. Curr. Opin. Chem. Biol. 15:267-275.

39. Chirpich TP, Zappia V, Costilow RN, Barker HA. 1970. Lysine 2,3-aminomutase. Purification and properties of a pyridoxal phosphate and S-adenosylmethionine-activated enzyme. J. Biol. Chem. 245:1778-1789.

40. Harder J, Eliasson R, Pontis E, Ballinger MD, Reichard P. 1992. Activation of the anaerobic ribonucleotide reductase from Escherichia coli by S-adenosylmethionine. J. Biol. Chem. 267:25548-25552.

41. Curatti L, Ludden PW, Rubio LM. 2006. NifB-dependent in vitro synthesis of the iron- molybdenum cofactor of nitrogenase. Proc. Natl. Acad. Sci. U.S.A 103:5297-5301.

42. Walker RD, Duerre JA. 1975. S-adenosylhomocysteine metabolism in various species. Can. J. Biochem. 53:312-319.

43. Halliday NM, Hardie KR, Williams P, Winzer K, Barrett DA. 2010. Quantitative liquid chromatography-tandem mass spectrometry profiling of activated methyl cycle metabolites involved in LuxS-dependent quorum sensing in Escherichia coli. Anal. Biochem. 403:20-29.

44. Bennett BD, Kimball EH, Gao M, Osterhout R, Van Dien SJ, Rabinowitz JD. 2009. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat. Chem. Biol. 5:593-599.

45. Kuramitsu HK, Udaka S, Moyed HS. 1964. Induction of inosine 5'-phosphate dehydrogenase and xanthosine 5'-phosphate aminase by ribosyl-4-amino-5- imidazolecarboxamide in purine-requiring mutants of Escherichia coli B. J. Biol. Chem. 239:3425-3430.

46. Moyed HS. 1964. Inhibition of the biosynthesis of the pyrimidine portion of thiamine by adenosine. J. Bacteriol. 88:1024-1029.

47. Shisler KA, Broderick JB. 2012. Emerging themes in radical SAM chemistry. Curr. Opin. Struct. Biol. 22:701-710.

CHAPTER 4

THE RHODANESE DOMAIN OF THII IS BOTH NECESSARY

AND SUFFICIENT FOR SYNTHESIS OF THE THIAZOLE MOIETY OF

THIAMINE IN SALMONELLA ENTERICA1

1 Palmer L. D.*, N. C. Martinez-Gomez*, E. Vivas, P. L. Roach, D. M. Downs. 2011. Journal of Bacteriology. 193(18):4582-7. *Contributed equally. Reprinted here with permission of publisher. 83

4.1 ABSTRACT

In Salmonella enterica, ThiI is a bifunctional enzyme required for the synthesis of both the 4-thiouridine modification in tRNA and the thiazole moiety of thiamine. In 4-thiouridine biosynthesis, ThiI adenylates the tRNA uridine and transfers sulfur from a persulfide formed on the protein. The role of ThiI in thiazole synthesis is not yet well understood. Mutational analysis described here found that ThiI residues required for 4-thiouridine synthesis were not involved in thiazole biosynthesis. The data further showed that the C-terminal rhodanese domain of ThiI was sufficient for thiazole synthesis in vivo. Together, these data support the conclusion that sulfur mobilization in thiazole synthesis is mechanistically distinct from that in 4-thiouridine synthesis and suggest that functional annotation of ThiI in genome sequences should be readdressed.

Nutritional studies described here identified an additional cysteine-dependent mechanism for sulfur mobilization to thiazole that did not require ThiI, IscS, SufS, or glutathione. The latter mechanism may provide insights into the chemistry used for sulfur mobilization to thiazole in organisms that do not utilize ThiI.

4.2 INTRODUCTION

Thiamine pyrophosphate (TPP) is an essential coenzyme used by pyruvate dehydrogenase, α- ketoacid decarboxylase, α-ketoacid oxidase transketolase, and other enzymes, where it stabilizes acyl carbanion intermediates through dissociation of the proton at C-2 of the thiazolium ring (9).

Though thiamine was isolated in the 1930s and its chemical synthesis was achieved 4 years later, significant questions remain about the biochemical details of TPP synthesis. Genetic studies, primarily in Salmonella enterica and Escherichia coli, have defined 10 gene products involved in

TPP synthesis in bacteria (Fig. 4.1) (10). With the exception of ThiI, a role for each of these gene products in TPP synthesis has been reconstituted in vitro (3, 7, 12, 13, 20, 24, 29). Microbes and 84

FIGURE 4.1 Thiamine biosynthesis in S. enterica. A schematic of the thiamine biosynthetic pathway in S. enterica is shown. The enzymes involved in each step are indicated above the relevant arrow. Abbre- viations: Cys, cysteine; DXP, deoxyxylulose phosphate; Tyr, tyrosine; AIR, 5-aminoimidazole ribotide; THZ-P, 2-carboxy-4-methyl-5-β -hydroxyethylthiazole phosphate; HMP-PP, 4-amino-5-hydroxymethyl-2- methylpyrimidine pyrophosphate.

85 plants synthesize thiamine de novo with separate pathways for the formation of the thiazole (2- carboxy-4-methyl-5-β-hydroxyethylthiazole phosphate [THZ-P]) and pyrimidine (4-amino-5- hydroxymethyl-2-methylpyrimidine pyrophosphate [HMP-PP]) moieties (1, 5, 18, 22). In bacteria, the two moieties are joined by thiamine phosphate synthase, the product of the thiE gene, to form thiamine monophosphate. The biologically active form of the coenzyme, thiamine pyrophosphate (TPP), is then generated by a single phosphorylation catalyzed by thiamine monophosphate kinase (ThiL) (14, 29).

The thiI locus was defined by lesions that caused a thiazole auxotrophy (28) and was subsequently found to be allelic with nuvA, a locus required for 4-thiouridine modification of tRNA (15). Synthesis of 4-thiouridine has been reconstituted in vitro, and our current understanding is summarized in Fig. 4.2. Experiments in vitro demonstrated the transfer of sulfur from the source cysteine via an IscS persulfide to form a ThiI persulfide (11) located at residue

Cys456 (19). In 4-thiouridine biosynthesis, ThiI activates the tRNA uridine as its adenylate (30,

32), which reacts with the Cys456 persulfide (11, 19). Residue Cys344 then acts as a nucleophile, forming a Cys456-Cys344 disulfide and releasing 4-thiouridine. The ThiI disulfide is then reduced by cellular reductants to complete the reaction cycle (17, 27). The ThiI protein in

E. coli/S. enterica contains three distinct domains, and each has been implicated in the mechanism of 4-thiouridine synthesis (Fig. 4.3). The THUMP domain is thought to bind the tRNA, the adenine nucleotide alpha-hydrolase domain (AANH) is responsible for uridine adenylation, and the C-terminal rhodanese domain participates in sulfur transfer (2, 6, 27, 32).

Five ThiI residues have been shown to be important for 4-thiouridine biosynthesis: Asp189,

Lys321, Cys207, Cys344, and Cys456. Asp189 and Lys321 are conserved residues in the AANH domain and are required for adenylation of the uridine (16). Cys207 is also in the AANH 86

FIGURE 4.2 Sulfur mobilization by ThiI in 4-thiouridine biosynthesis. Events depicted occur in the active site of ThiI. The boxes represent the THUMP and AANH domains of ThiI, while the gray circles represent the rhodanese domain. Critical cysteine residues 344 and 456 are shown. tRNA represents the rest of the tRNA molecule containing the uridine being acted on by ThiI. 87 domain, and the ThiIC207A variant showed intermediate activity in vivo and in vitro (17). Cys456 is conserved in the rhodanese domain and carries the nucleophilic cysteine persulfide that is essential for the transfer of sulfur to uridine (19). Cys344 is a highly conserved residue located in the AANH domain that forms a disulfide bond with Cys456, releasing 4-thiouridine (17).

Sulfur mobilization in thiazole synthesis shares many features with those described above for 4-thiouridine synthesis. In thiazole synthesis, ThiFS forms a stable complex, and ThiF catalyzes the adenylation of the terminal glycine residue of ThiS (31). It is the acyladenylate on

ThiS that accepts sulfur from a persulfide donor (24). It has been suggested that sulfur mobilization in thiazole synthesis utilizes ThiI as a persulfide carrier between IscS and the ThiS- acyladenylate (17, 24). Indeed, in vivo results suggest that thiI is required for the formation of the

ThiS thiocarboxylate (24), although ThiI was not required for the in vitro formation of the ThiS thiocarboxylate (31). The role of ThiI in sulfur mobilization in thiazole synthesis has not been directly tested by mutational analyses or in vitro studies.

This study was initiated to probe the role of ThiI in thiazole synthesis in the context of what is known about sulfur mobilization in 4-thiouridine biosynthesis. The results here showed that ThiI residues critical for 4-thiouridine biosynthesis were not required for thiazole synthesis and that the rhodanese domain of ThiI was sufficient for synthesis of thiazole in vivo. Nutritional studies showed that there was an alternative mechanism of sulfur mobilization to thiazole that depends on exogenous cysteine. These results provide insights into sulfur mobilization in thiazole synthesis and described important criteria for the appropriate annotation of ThiI in genomic data. 88

4.3 MATERIALS AND METHODS

Strains, media, and chemicals. Minimal medium was NCE (26) supplemented with MgSO4 (1 mM), trace minerals (0.1×) (4), and glucose (11 mM). The final concentration of thiamine or thiazole was 100 nM. Rich medium was Difco nutrient broth (NB; 8 g/liter) with NaCl (5 g/liter).

Solid medium contained 1.5% agar. Antibiotics were added at the following concentrations in rich and minimal media, respectively: chloramphenicol (Cm), 20 and 5 g/liter; ampicillin (Ap),

150 and 30 g/liter; tetracycline (Tc), 20 and 10 g/liter; kanamycin (Km), 50 and 12.5 g/liter. All chemicals were purchased from Sigma-Aldrich, St. Louis, MO. The strains used in this study were derivatives of S. enterica strain LT2, and their genotypes are listed in Table 4.1. All strains were generated for this study or were part of the laboratory collection.

Mutant construction. (i) Local mutagenesis. A P22 lysate (HT int201 [21, 23]) was generated on strain DM12255, which contains a Tn10d (Tc) insertion in phnU, linked to the thiI locus. This lysate was mutagenized by hydroxylamine (8) and used to transduce a wild-type strain

(DM10000) to Tcr. Tcr transductants were screened for those that required exogenous thiamine when grown on glucose minimal medium (Thi−). The causative mutations were transduced into a new genetic background to confirm causation of the thiamine requirement, and the lesions in thiI were identified by sequence analysis. All DNA sequencing was performed at the University of

Wisconsin—Madison Biotechnology Center.

(ii) Site-directed mutagenesis. A plasmid (pDM1322) containing thiI ligated into the NdeI and

BamHI sites of pET-14b (Novagen) was constructed. Relevant thiI mutants were generated by site-directed mutagenesis with the QuikChange XL site-directed mutagenesis kit according to the manufacturer's protocol (Agilent Technologies, Santa Clara, CA). Sequences of primers used are 89

TABLE 4.1 Strain and plasmid list.

Strain Genotype DM10000 Wild type DM269 thiI887::Tn10d(Tc) DM11537 ∆thiF DM12225 phnU::Tn10d(Tc) DM12260 thiI887::Tn10d(Tc) pET-14b DM11774 thiI887::Tn10d(Tc) pDM1322 (pET-ThiI) DM11544 thiI887::Tn10d(Tc) pDM1324 (pET-ThiID189A) DM11545 thiI887::Tn10d(Tc) pDM1326 (pET-ThiIC207S) DM11546 thiI887::Tn10d(Tc) pDM1327 (pET-ThiIK321M) DM11528 thiI887::Tn10d(Tc) pDM1328 (pET-ThiIC344S) DM11547 thiI887::Tn10d(Tc) pDM1329 (pET-ThiIC456S) DM13436 thiI887::Tn10d(Tc) pSU19 DM13437 thiI887::Tn10d(Tc) pDM1319 (pSUthiI) DM13438 thiI887::Tn10d(Tc) pDM1320 (pSUthiI-Cter) DM13439 thiI887::Tn10d(Tc) pDM1321 (pSUthiI-CterC456S) DM13477 thiI887::Tn10d(Tc) pDM1331 (pSUthiI-CterR414H) DM10959 thiI1205::Tn10d(Tc) DM13162 ∆iscS DM13304 thiI1205::Tn10d(Tc) ∆iscS DM13483 thiI1205::Tn10d(Tc) gshA102::MudJ(Kn) DM13484 thiI1205::Tn10d(Tc) sufS::Cat Plasmid Vector; Protein encoded pDM1322 (pET-ThiI) pET-14b (ApR); ThiI pDM1324 (pET-ThiID189A) pET-14b (ApR); ThiID189A pDM1326 (pET-ThiIC207S) pET-14b (ApR); ThiIC207S pDM1327 (pET-ThiIK321M) pET-14b (ApR); ThiIK321M pDM1328 (pET-ThiIC344S) pET-14b (ApR); ThiIC344S pDM1329 (pET-ThiIC456S) pET-14b (ApR); ThiIC456S pDM1318 (pSCthiI) pSC-B (ApR, KnR); ThiI pDM1316 (pSCthiI-Cter) pSC-B (ApR, KnR); ThiI-Cter (residues 385-482) pDM1317 (pSCthiI-CterC456S) pSC-B (ApR, KnR); ThiI-CterC456S (residues 385-482) pDM1330 (pSCthiI-CterR414H) pSC-B (ApR, KnR); ThiI-CterR414H (residues 385-482) pDM1319 (pSUthiI) pSU19 (CmR); ThiI pDM1320 (pSUthiI-Cter) pSU19 (CmR); ThiI-Cter (residues 385-482) pDM1321 (pSUthiI-CterC456S) pSU19 (CmR); ThiI-CterC456S (residues 385-482) pDM1331 (pSCthiI-CterR414H) pSU19 (CmR); ThiI-CterR414H (residues 385-482) 90

available from the corresponding author upon request. The presence of the relevant alleles in the resulting constructs was confirmed by sequence analysis.

Expression of the C-terminal domain of ThiI. The C-terminal domain of ThiI (residues 385 to

482) was amplified by PCR using chromosomal DNA of wild-type S. enterica as a template and using primers SalthiI-R385for and SalthiI-endrev. The primers introduced an initiating methionine and restriction sites (BamHI and HindIII) to the amplified sequence. The amplified

DNA was cloned as a blunt-end fragment into plasmid pSC-B-amp/kan (Stratagene) to give plasmid pDM1316. Plasmid pDM1316 was digested with BamHI and HindIII at the sites engineered in the primers, and the resulting 312-bp C-terminal thiI fragment was cloned into pSU19 to give plasmid pDM1320. Plasmids containing the C-terminal variants, C456S and

R414H, were constructed by cloning a PCR amplification product from pDM1329 DNA or

DM12302 chromosomal DNA, respectively. In each case, the constructs were confirmed by sequence analyses.

Nutritional analysis. Nutritional requirements were determined by liquid growth analysis and in solid medium by using soft agar overlays. For liquid growth analysis, strains were grown overnight in NB (antibiotic was present for plasmid-containing strains). Cells were pelleted and resuspended in NaCl (0.85%). A 0.2-ml aliquot of this suspension was used to inoculate 5 ml minimal medium containing glucose. Cell growth was measured by monitoring absorbance at

650 nm (A650) while strains were incubated with shaking (200 rpm) at 37°C. Alternatively, 5 µl of cell suspension was used to inoculate 195 µl of the appropriate minimal medium in each well of a 96-well microtiter plate. Growth at 37°C with shaking at intensity level 2 was monitored using a microplate spectrophotometer (Bio-Tek Instruments). The specific growth rate was determined as µ = ln(X/X0)/T, where X is the A650 value during the exponentially linear portion of 91

FIGURE 4.3 Structural domains of S. enterica ThiI. Motifs in the S. enterica ThiI are based on those suggested previously (27), and the residue numbers are those of E. coli. Residues 1 to 163 contain an N-terminal ferredoxin-like (NFLD) domain and a THUMP domain (2), both of which are thought to interact with the tRNA (27). Resi- dues 173 to 400 contain the adenylation domain (AANH). The C-terminal portion contains the rhodanese-like domain. 92

growth (routinely between A650 0.2 and 0.7) and T is time in hours. For soft agar overlays, cells were grown overnight in NB. An 0.2-ml aliquot of the overnight culture was used to seed 3.5 ml of 0.75% agar, which was overlaid on a minimal medium plate. Nutritional supplements were then spotted, and growth was assessed after 18 h of incubation at 37°C.

Bioinformatic analysis. The TIGRfam TIGR00342 was downloaded from the Comprehensive

Microbial Resource (CMR; http://cmr.jcvi.org), using the CMR Gene Attribute Download tool and the JCVI/CMR's OMNIOME.pep database. Using CMR, blastp (WU-BLAST version

2.0MP) was run on TIGR00342 using E. coli K-12 ThiI residues 404 to 482 as the query. Default parameters were used, except that the E-value cutoff was lowered to 0.001. Members of

TIGR00342 containing the rhodanese domain were separated from those without a rhodanese domain, and duplicate species entries were removed. Bioinformatic analysis was performed by the Advanced Genome Analysis Resource facility at the University of Wisconsin Biotechnology

Center.

4.4 RESULTS AND DISCUSSION

The functions of ThiI in 4-thiouridine biosynthesis and thiamine synthesis are genetically separable. Five residues of ThiI have been shown to be important for synthesis of the 4- thiouridine modification of tRNA in vitro (Fig. 4.3) (16, 17, 19). These residues were independently changed in S. enterica ThiI, and plasmids carrying each of the variants (C344S,

C456S, D189A, K321M, and C207S) or the wild-type gene were constructed. The resulting derivatives of plasmid pET-14b expressing ThiI proteins were transformed into a mutant strain lacking thiI, DM269 (thiI887::Tn10d). In the resulting strains, the plasmid-encoded protein was the only source of ThiI. Each strain was monitored for growth in the absence of thiamine to assess complementation of the thiI lesion. All variants except ThiIC456S restored growth of a thiI 93

FIGURE 4.4 ThiI variants defective in 4-thiouridine biosynthesis complement a thiI mutant strain. Growth of strains at 37°C with shaking was monitored over time as absorbance at 650 nm. Strains were grown in minimal NCE medium with glucose and ampicillin, all strains grew with wild-type rate and yield if thiamine was added to the medium. Strains were DM269 (thiI887::Tn10d) with plasmids pET-14b (□), pET-ThiI (○), pET-ThiIC344S (), pET-ThiID189A (▼), and pET-ThiIC456S (▲). Data shown are the average and standard deviation of three independent cultures. 94 mutant strain in minimal medium. Some of these data are shown in Fig. 4.4. These results indicated that of the five ThiI residues required for 4-thiouridine synthesis, only residue Cys456 was required for thiazole synthesis.

The ThiI variants that complemented a thiI mutant strain provided insights into the mechanism of ThiI in thiazole biosynthesis. ThiI residues Asp189, Lys321, and Cys207 lie in the adenylation domain. Substitution at any of these residues eliminated or decreased tRNA modification, consistent with the role of ThiI in adenylating the tRNA uridine in synthesis of the

4-thiouridine (17). In contrast, ThiI variants with substitutions at any of these residues were proficient for thiazole synthesis in vivo. These data allow the conclusion that the adenylation activity of ThiI is not required for thiazole synthesis. Similarly, residues Cys456 and Cys344 were directly involved in sulfur mobilization for the 4-thiouridine modification of tRNA in vitro

(17, 19). In contrast, Cys456, but not Cys344, was required for thiazole synthesis in vivo. These data support the conclusion that sulfur mobilization in thiazole synthesis is mechanistically distinct from that in 4-thiouridine synthesis.

The C-terminal rhodanese domain of ThiI is sufficient for thiazole synthesis. Local mutagenesis of the thiI gene generated 16 independent mutants that required thiamine for growth. The distribution of causative mutations was striking, with 14 of the 16 mutations generating a nonsense codon, two of which were isolated twice (Table 4.2). All of the lesions affected the rhodanese domain (residues 385 to 482). Of the two missense mutations isolated, one caused a substitution at residue Cys456 (C456Y) and the other resulted in an R414H substitution in the rhodanese domain.

The above results raised the question of whether the rhodanese domain of ThiI was sufficient for synthesis of thiazole. An algorithm to detect linker regions in proteins defined such 95

TABLE 4.2 ThiI variants that result in a thiamine requirement in vivo.

Strain Allele Base change(s) Protein Varianta DM12300 thiI1209 C1172T, C1414T T391I, Q472* DM12302 thiI1210 G1241A R414H DM12304 thiI1211 C1342T Q448* DM12355 thiI1212 C1255T Q419* DM12357 thiI1213 C1255T Q419* DM12359 thiI1214 G1356A W452* DM12361 thiI1215 C886T Q296* DM12422 thiI1216 C175T Q59* DM12424 thiI1217 G1397A C456Y DM12426 thiI1218 C1342T Q448* DM12428 thiI1219 C1393T Q465* DM12430 thiI1220 C1342T Q448* DM12432 thiI1221 C1342T Q448* DM12883 thiI1222 C886T Q296* DM12885 thiI1223 G1355A W452* DM12889 thiI1224 C886T Q296* aAsterisks indicate the generation of nonsense codons.

96 a region in ThiI from residues 381 to 390 (25). A plasmid expressing only the rhodanese domain of ThiI (residues 385 to 482) was constructed and used to test complementation of a thiI mutant.

This construct was sufficient to restore thiamine-independent growth of a thiI mutant strain (4.

5). As shown in Fig. 4.5, complementation by the rhodanese domain was eliminated when residue Cys456 was changed to Ser, or when Arg414 was changed to His. Together, these results support the hypothesis that the rhodanese domain of ThiI was sufficient for thiazole synthesis.

These data further demonstrate that ThiI mediated sulfur transfer in 4-thouridine synthesis via a different mechanism than that in the synthesis of thiazole.

A possible role for the rhodanese domain in sulfur mobilization to thiazole. The accepted mechanistic model for the role of ThiI in 4-thiouridine synthesis is shown in Fig. 4.2. In this model, the THUMP domain binds the tRNA and the AANH domain activates the tRNA uridine by adenylation (11, 16, 32). The ThiI persulfide on Cys456 then displaces the adenosyl group on the uridine (19, 30). Finally, Cys344 acts as a nucleophile, forming a Cys456-Cys344 disulfide and releasing the product 4-thiouridine. The resulting disulfide in ThiI can then be reduced to begin the next reaction cycle (17, 27, 30). Previously, it was proposed that ThiI served a similar role in sulfur transfer in thiazole synthesis (17, 24). Figure 4.6 represents a mechanism for ThiI in sulfur mobilization to thiazole that is both consistent with previous studies and supported by the data described above. The mechanism depicted implies analogous chemistry in 4-thiouridine biosynthesis and thiazole biosynthesis. In both cases, the substrate (i.e., tRNA uridine or ThiS terminal glycine) is first activated by adenylation. In the case of 4-thiouridine biosynthesis, ThiI catalyzes the substrate adenylation reaction. In thiazole synthesis, ThiF has been shown to adenylate the terminal glycine residue of ThiS, generating an acyladenylate (24). The data showing that the adenylation activity of ThiI is not required for thiazole synthesis support this 97

FIGURE 4.5 The rhodanese domain of ThiI is sufficient for thiamine-independent growth. Strains were grown in NCE medium supplemented with glucose. The following plasmids were introduced into DM269 (thiI): pSU19 (□), pSUThiI (○), pSUThiI-Cter (▼), pSUThiI-CterC456S (▲), and pSUThiI-CterR414H (). Strain DM269 with pSU19 was also grown in medium supplemented with thiamine (■).

98 assignment. In both pathways, IscS transfers a sulfur atom to ThiI Cys456, forming a persulfide, a proposal consistent with the requirement for Cys456 in both pathways. This persulfide then attacks the substrate (uridine or ThiS) adenylate, displacing the adenosyl group and forming a disulfide between ThiI and the 4 position of uridine or an acyl persulfide with the C terminus of

ThiS. In the final step, a second cysteine (ThiI Cys344 or ThiF Cys184, respectively) attacks the disulfide bond to release either the final product (4-thiouridine biosynthesis) or ThiI (in the case of thiazole biosynthesis). The last step is consistent with the in vivo data showing that an E. coli strain with a ThiFC184S variant is a thiamine auxotroph (31), which implicated the ThiFS acylpersulfide as either an intermediate or a final sulfur donor in thiazole synthesis (24, 31).

Evaluation of ThiI annotations in genome sequences. Search of the JCVI/CMR's

OMNIOME.pep database for proteins with the TIGR00342 multidomain (THUMP and AANH) identified 207 proteins annotated as thiamine biosynthesis/4-thiouridine biosynthesis proteins. Of these, 58 (28%) contained a rhodanese domain, exemplified by the E. coli/S. enterica-type ThiI discussed herein. However, 159 (72%) had only the THUMP and AANH domains, exemplified by Bacillus subtilis. The THUMP domain, the AANH domain, and the rhodanese domain are implicated in the synthesis of 4-thiouridine (2, 16, 17). However, data in the above sections showed that the rhodanese domain alone is sufficient for ThiI to function in thiazole synthesis.

The B. subtilis ThiI is among the 72% of the TIGR00342-containing proteins without a rhodanese domain annotated as thiamine biosynthetic proteins, and yet this protein is not required for thiazole synthesis (J. Perkins, personal communication; P. Dos Santos, personal communication). A role for the ThiI protein of B. subtilis in the synthesis of 4-thiouridine is presumed by the annotation but has not been reported.

FIGURE 4.6 Proposed model for ThiI mechanism in thiazole biosynthesis. Sulfur mobilization is proposed to proceed as described in the text. The gray circle represents the ThiI rhodanese domain. The terminal glycine carboxyl group of ThiS is depicted. 100

S. enterica ThiI is a bifunctional enzyme that (i) binds and adenylates tRNA and (ii) mobilizes sulfur. Both functions are required for 4-thiouridine biosynthesis, while only sulfur mobilization is required for thiazole synthesis, as shown here. These findings suggest that the current annotation of the ThiI family of proteins should be revisited to better reflect their separable roles in the synthesis of 4-thiouridine and thiazole synthesis. Specifically, proteins annotated as ThiI based on containing only the AANH and THUMP domains are unlikely to be involved in thiazole synthesis and would be better annotated as 4-thiouridine sulfurtransferases.

Based on the results here, only those proteins with the THUMP, AANH, and rhodanese domain are likely to be involved in both pathways. At this point, it is unclear how organisms with ThiI proteins with only two domains mobilize sulfur for 4-thiouridine synthesis in vivo. It has been suggested that an orphan rhodanese domain acts with the ThiI protein in B. subtilis and other organisms that contain a short ThiI protein (27).

Exogenous cysteine can restore thiamine-independent growth of a thiI mutant strain. The result that the rhodanese domain of ThiI was sufficient for thiazole synthesis, coupled with the fact that not all thiamine-producing organisms have a rhodanese domain associated with a ThiI enzyme, suggested that another paradigm(s) for sulfur mobilization to thiazole synthesis might exist. For instance, the IscS persulfide could serve as the direct sulfur donor in some metabolic networks, or organisms may employ a different mechanism altogether. It seemed plausible that both the redox environment of the cell and the intracellular concentration of cysteine or other free thiols could influence these possibilities, perhaps in organismal or environmentally specific ways. This idea is supported by the observation that in vitro cysteine and IscS could serve as the sole sulfur source to ThiS (12). Further, sulfide can provide the source of sulfur in vitro without 101

ThiI, ThiF, or ThiS (12, 20) (N. C. Martinez-Gomez, unpublished data), suggesting that thiazole synthase, ThiG, could utilize a variety of sulfides.

In vivo, a thiI mutant requires thiazole, and efforts to isolate suppressor mutations by spontaneous or chemical mutagenesis were unsuccessful. As might be predicted from the in vitro results, sulfide provided in the form of Li2S or Na2S restored growth of a thiI mutant strain in the absence of thiamine, although this result was not considered physiologically relevant (data not shown). Unexpectedly, nutrient studies found that addition of cysteine (0.2 mM) to the medium allowed growth of a thiI mutant strain in the absence of thiazole (Table 4.3). Restoration of thiamine-independent growth was specific to the addition of cysteine; no other physiologically relevant sulfur source or reducing compound tested restored growth to a thiI mutant strain (data not shown). Significantly, the ability of cysteine to allow growth did not require IscS. Neither did this effect of cysteine require SufS (a cysteine desulfurase) or GshA (glutathione biosynthetic enzyme). These conclusions were based on the data in Table 4.3, which showed that the thiazole requirement of each double mutant strain (thiI iscS, thiI gshA, and thiI sufS) was satisfied by exogenous cysteine. The slightly reduced growth rate of the double mutants compared to the thiI mutant is consistent with the deleterious effects of these mutations on other aspects of metabolism. Significantly, the addition of cysteine fully satisfied the thiazole requirement as reflected by the similar growth rates allowed by cysteine and thiazole. Finally, cysteine did not allow growth of a thiF mutant in the absence of thiazole (data not shown), indicating that cysteine did not serve as a direct sulfur source to ThiG. Together, these data suggest that an alternative mechanism for sulfur mobilization to ThiS existed in the cell that depends on exogenous cysteine. The finding that cysteine satisfied the thiazole requirement of iscS, gshA, 102

TABLE 4.3 Cysteine allows growth of a thiI mutant in minimal medium. Strains were grown in NCE medium supplemented with glucose (11 mM) and the indicated additions. Growth rate is reported as µ (µ=ln(X/Xo)/T ) and final cell yield is A650 after 25 hours of growth. Data shown are the average and standard deviation of three independent cultures. Specific growth rate Final cell yield Strain Relevant Min Cys THZ Min Cys THZ genotype DM10959 thiI 0.35±0.01 0.74±0.03 0.78±0.04 0.45±0.09 1.45±0.03 1.56±0.00 DM13304 thiI iscS 0.12±0.01 0.31±0.01 0.36±0.01 0.36±0.03 0.98±0.03 1.45±0.02 DM13483 thiI gshA 0.27±0.01 0.50±0.01 0.53±0.01 0.39±0.02 1.44±0.00 1.53±0.04 DM13484 thiI sufS 0.30±0.01 0.69±0.07 0.69±0.01 0.43±0.01 1.45±0.04 1.56±0.02 103 and sufS derivatives of a thiI mutant indicates that the alternative mechanism of sulfur mobilization does not depend on either of the dominant cysteine desulfurases in the cell or the primary thiol reductant, glutathione. It is possible that an additional cysteine desulfurase, possibly via an orphan rhodanese in the cell, can mediate sulfur transfer to ThiS. The requirement for exogenous cysteine might indicate a need to accumulate an appropriate persulfide. Such a scenario may reflect and/or mimic a paradigm for thiazole synthesis in organisms that do not require ThiI for thiazole biosynthesis.

Conclusions. In S. enterica and E. coli, ThiI is involved in the synthesis of both thiazole and 4- thiouridine (15, 28). The data here showed that the THUMP and AANH domains of this protein, which are the basis for annotation of proteins as ThiI in genomes, are not required for thiazole synthesis. This work emphasizes that caveats can exist when annotation of gene function is propagated from a single experimental organism. Finding an additional cysteine-dependent mechanism of sulfur mobilization to thiazole in vivo suggests that there may be diverse processes for sulfur mobilization to thiazole. Further characterization of this mechanism could improve understanding of the relationship between the thiol environment in the cell and mobilization of sulfur to thiazole synthesis.

4.5 ACKNOWLEDGMENTS

We acknowledge undergraduates James Vasta and Patrick Borchert for isolating several of the thiI mutants. The National Institutes of Health grant GM47296 and the 21st Century

Scientists Scholars program from the J. M. McDonnell Foundation to D.M.D. supported this work. L.D.P. was supported by NSF through Graduate Research Fellowship grant DGE-

0718123. 104

N. Cecilia Martinez-Gomez first identified the cysteine phenotype, constructed the parent pET14b-ThiI construct and wrote much of the manuscript. E. Ignacio Vivas constructed the strains and performed the experiment depicted in Figure 4.5 and constructed the strains characterized in Table 4.3. Peter L. Roach contributed to Figure 4.2 and Figure 4.6. James Vasta and Patrick Borchert isolated and sequenced the thiI mutant alleles under my direction. I constructed the remaining strains, performed the remaining experiments, generated the figures and contributed to writing.

4.6 REFERENCES

1. Ajjawi, I., M. A. Rodriguez Milla, J. Cushman, and D. K. Shintani. 2007. Thiamin pyrophosphokinase is required for thiamin cofactor activation in Arabidopsis. Plant Mol. Biol. 65:151–162.

2. Aravind, L., and E. V. Koonin. 2001. THUMP—a predicted RNA-binding domain shared by 4-thiouridine, pseudouridine synthases and RNA methylases. Trends Biochem. Sci. 26:215–217.

3. Backstrom, A. D., R. Austin, S. McMordie, and T. P. Begley. 1995. Biosynthesis of thiamin. I. The function of the thiE gene product. J. Am. Chem. Soc. 117:2351–2352.

4. Balch, W. E., G. E. Fox, L. J. Magrum, C. R. Woese, and R. S. Wolfe. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260–296.

5. Begley, T. P., et al. 1999. Thiamin biosynthesis in prokaryotes. Arch. Micro- biol. 171:293–300.

6. Bordo, D., and P. Bork. 2002. The rhodanese/Cdc25 phosphatase superfamily. Sequence- structure-function relations. EMBO Rep. 3:741–746.

7. Chatterjee, A., et al. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758– 765.

8. Hong, J. S., and B. N. Ames. 1971. Localized mutagenesis of any specific small region of the bacterial chromosome. Proc. Natl. Acad. Sci. U. S. A. 68:3158–3162.

9. Jordan, F. 2003. Current mechanistic understanding of thiamin diphosphate-dependent enzymatic reactions. Nat. Prod. Rep. 20:184–201.

10. Jurgenson, C. T., T. P. Begley, and S. E. Ealick. 2009. The structural and biochemical foundations of thiamin biosynthesis. Annu. Rev. Biochem. 78: 569–603. 105

11. Kambampati, R., and C. T. Lauhon. 2000. Evidence for the transfer of sulfane sulfur from IscS to ThiI during the in vitro biosynthesis of 4-thiouridine in Escherichia coli tRNA. J. Biol. Chem. 275:10727–10730.

12. Kriek, M., et al. 2007. Thiazole synthase from Escherichia coli: an investigation of the substrates and purified proteins required for activity in vitro. J. Biol. Chem. 282:17413– 17423.

13. Martinez-Gomez, N. C., and D. M. Downs. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2- methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054–9056.

14. McCulloch, K. M., C. Kinsland, T. P. Begley, and S. E. Ealick. 2008. Structural studies of thiamin monophosphate kinase in complex with substrates and products. Biochemistry 47:3810–3821.

15. Mueller, E. G., C. J. Buck, P. M. Palenchar, L. E. Barnhart, and J. L. Paulson. 1998. Identification of a gene involved in the generation of 4-thiouridine in tRNA. Nucleic Acids Res. 26:2606–2610.

16. Mueller, E. G., and P. M. Palenchar. 1999. Using genomic information to investigate the function of ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis. Protein Sci. 8:2424–2427.

17. Mueller, E. G., P. M. Palenchar, and C. J. Buck. 2001. The role of the cysteine residues of ThiI in the generation of 4-thiouridine in tRNA. J. Biol. Chem. 276:33588–33595.

18. Nosaka, K. 2006. Recent progress in understanding thiamin biosynthesis and its genetic regulation in Saccharomyces cerevisiae. Appl. Microbiol. Biotech- nol. 72:30–40.

19. Palenchar, P. M., C. J. Buck, H. Cheng, T. J. Larson, and E. G. Mueller. 2000. Evidence that ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis, may be a sulfurtransferase that proceeds through a persulfide intermediate. J. Biol. Chem. 275:8283–8286.

20. Park, J. H., et al. 2003. Biosynthesis of the thiazole moiety of thiamin pyrophosphate (vitamin B1). Biochemistry 42:12430–12438.

21. Roberts, G. P. 1978. Isolation and characterization of informational suppressors in Salmonella typhimurium. Ph.D. thesis. University of California, Berkeley, CA.

22. Roje, S. 2007. Vitamin B biosynthesis in plants. Phytochemistry 68:1904– 1921.

23. Schmieger, H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75–88.

24. Taylor, S. V., et al. 1998. Thiamin biosynthesis in Escherichia coli. Identification of ThiS thiocarboxylate as the immediate sulfur donor in the thiazole formation. J. Biol. Chem. 106

273:16555–16560.

25. Udwary, D. W., M. Merski, and C. A. Townsend. 2002. A method for prediction of the locations of linker regions within large multifunctional proteins, and application to a type I polyketide synthase. J. Mol. Biol. 323:585-598.

26. Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218:97–106.

27. Waterman, D. G., M. Ortiz-Lombardia, M. J. Fogg, E. V. Koonin, and A. A. Antson. 2006. Crystal structure of Bacillus anthracis ThiI, a tRNA-modifying enzyme containing the predicted RNA-binding THUMP domain. J. Mol. Biol. 356:97–110.

28. Webb, E., K. Claas, and D. M. Downs. 1997. Characterization of thiI, a new gene involved in thiazole biosynthesis in Salmonella typhimurium. J. Bacteriol. 179:4399– 4402.

29. Webb, E., and D. Downs. 1997. Characterization of thiL, encoding thiamin- monophosphate kinase, in Salmonella typhimurium. J. Biol. Chem. 272: 15702–15707.

30. Wright, C. M., P. M. Palenchar, and E. G. Mueller. 2002. A paradigm for biological sulfur transfers via persulfide groups: a persulfide-disulfide-thiol cycle in 4-thiouridine biosynthesis. Chem. Commun. (Camb.) 2002:2708– 2709.

31. Xi, J., Y. Ge, C. Kinsland, F. W. McLafferty, and T. P. Begley. 2001. Biosynthesis of the thiazole moiety of thiamin in Escherichia coli: identification of an acyldisulfide-linked protein-protein conjugate that is functionally analogous to the ubiquitin/E1 complex. Proc. Natl. Acad. Sci. U. S. A. 98:8513– 8518.

32. You, D., T. Xu, F. Yao, X. Zhou, and Z. Deng. 2008. Direct evidence that ThiI is an ATP pyrophosphatase for the adenylation of uridine in 4-thiouridine biosynthesis. Chembiochem 9:1879–1882.

107

CHAPTER 5

REDUNDANCY IN SULFUR TRAFFICKING TO THIAMINE BIOSYNTHESIS

IN SALMONELLA ENTERICA1

1 Palmer L.D., M.H. Leung, D.M. Downs. To be submitted to Journal of Bacteriology.

108

5.1 ABSTRACT

Thiamine pyrophosphate is a required coenzyme that contains a mechanistically important sulfur atom. In Salmonella, sulfur is trafficked to both thiamine biosynthesis and 4- thiouridine biosynthesis by the enzyme ThiI using persulfide (R-S-S-H) chemistry. It was previously reported that a thiI mutant strain could grow independent of thiamine addition in the presence of cysteine, suggesting an alternative strategy for sulfur mobilization existed. Data reported here suggest that oxidation products of cysteine rescue growth of a thiI mutant strain.

The alternative sulfur mobilization requires the transporter YdjN and the cysteine desulfurase

CdsH. The data are consistent with a model in which sulfide produced by CdsH reacts with

- cystine (Cys-S-S-Cys), S-sulfocysteine (Cys-S-SO3 ) or another disulfide to form a small- molecule persulfide (R-S-S-H). The persulfide could then replace ThiI by donating sulfur to the thiamine sulfur carrier protein ThiS. This model describes a mechanism that might be exploited by organisms that lack ThiI but are capable of thiamine biosynthesis.

5.2 INTRODUCTION

Thiamine pyrophosphate (TPP) is a required cofactor for central metabolic enzymes in all domains of life, including pyruvate dehydrogenase, α-ketoglutarate dehydrogenase and transketolase. Humans require dietary thiamine, while many bacteria, plants and fungi are capable of synthesizing it de novo. Thiamine is composed of two independently synthesized moieties: 2-methyl-4-amino-5-hydroxymethylpyrimidine phosphate (HMP-P) and 5-(2-

Hydroxyethyl)-4-methylthiazole phosphate (THZ-P). In bacteria and plants, HMP-PP biosynthesis requires two dedicated enzymes. ThiC produces HMP-P from the purine biosynthetic intermediate 4-aminoimidazole ribotide (1, 2), which is then phosphorylated to

HMP-PP by ThiD (3, 4). In contrast, bacterial THZ-P biosynthesis requires multiple enzymes

109 and different phyla employ distinct strategies. Facultative anaerobes, such as the γ-proteobacteria

Salmonella enterica and Escherichia coli, use the enzymes ThiH, ThiG, ThiF, ThiS, ThiI, and

IscS (5-7). Other organisms, including Bacillus subtilis, have the glycine oxidase ThiO instead of

ThiH and do not require ThiI or IscS (8, 9). In all cases, ThiE ligates the HMP-PP and THZ-P moieties to form thiamine monophosphate, which is then phosphorylated to TPP by ThiL (10,

11).

The sulfur atom in THZ-P is required for TPP coenzyme activity because it holds a partial positive charge, stabilizing the carbanionic form that is required for α-decarboxylation and transketolation (12). In Salmonella enterica and E. coli, ThiI is a dual-function enzyme required for sulfur activation and mobilization to THZ-P biosynthesis and the 4-thiouridine modification of tRNA. A thiI mutant was defined by its requirement for THZ in Salmonella enterica (6). The thiI locus was found to be allelic with nuvA, a gene required for 4-thiouridine

(s4U) biosynthesis and near-UV sensitivity (13, 14); ThiI/NuvA thiolates base position 8 in tRNA to 4-thiouridine, which can act as a photosensor.

In S. enterica, ThiI contains multiple domains, including the N-terminal ferridoxin-like domain, the tRNA-binding THUMP domain, the adenine nucleotide α-hydrolase domain

(AANH, also referred to as the PP-loop pyrophosphatase domain) capable of adenylylation, and a C-terminal rhodanese-like domain (15-20). While all domains appear to be required for 4- thiouridine formation, only the C-terminal rhodanese-like domain was required for thiamine biosynthesis in Salmonella (18). Subsequently, “short” ThiI proteins that lack the C-terminal rhodanese domain were shown to be dispensable for THZ-P biosynthesis in the gram-positive

Firmicute B. subtilis and the archaeon Methanococcus maripaludis (9, 21).

110

In thiamine biosynthesis, ThiF adenylates the C-terminal glycine of the sulfur carrier enzyme ThiS (22, 23). An IscS cysteine persulfide (R-S-S-H) then donates its terminal sulfur to form a Cys456 persulfide in the ThiI rhodanese-like domain (24, 25). The Cys456 persulfide is hypothesized to then displace the adenyl group on ThiS, forming a ThiI Cys456-ThiS thiocarboxylate disulfide intermediate (18). The ThiF Cys184 would then displace ThiI, forming a ThiF Cys184-ThiS thiocarboxylate disulfide that serves as the sulfur donor to thiazole synthase

ThiG or is reduced to allow the ThiS thiocarboxylate to serve as the ultimate sulfur donor (23).

The dependence of this pathway on persulfide chemistry is a general paradigm of sulfur mobilization, with persulfides considered to be the primary sulfur donors to nucleosides and cofactors including lipoic acid, biotin and iron sulfur clusters (26).

Although a mechanism for ThiI sulfur donation to THZ-P in S. enterica and E. coli can be envisioned, the report that the thiamine requirement of a thiI mutant was rescued by exogenous cysteine (18) prompted this study to characterize the mechanism of ThiI-independent sulfur trafficking to thiamine.

5.3 MATERIALS AND METHODS

Strains, media and chemicals. Minimal medium was No Carbon E medium (NCE) (27) supplemented with MgSO4 (1 mM), trace minerals (0.1X) (28) and glucose (11 mM). When present, nutrients were added to the following concentrations: thiamine, 100 nM; cysteine, cystine, S-sulfocysteine, or Na2S, 100 µM. Cystine was dissolved in 0.1 N HCl to a 0.5 M stock solution before appropriate dilution. Rich media was Difco nutrient broth (NB; 8 g/L) with NaCl

(5 g/L) or superbroth (SB; tryptone (32 g/L), yeast extract (20 g/L), NaCl (5 g/L) with NaOH

(0.05 N)). Solid media contained 1.5% agar. Antibiotics were added at the following concentrations in rich and minimal media, respectively: chloramphenicol (Cm), 20 mg/L, 5

111 mg/L; ampicillin (Ap), 150 mg/L, 30 mg/L; tetracycline (Tc), 20 mg/L, 10 mg/L; kanamycin

(Kn), 50 mg/L, 150 mg/L. All chemicals were purchased from Sigma-Aldrich, St Louis, MO.

The strains used in this study were derivatives of S. enterica strain LT2 and were generated for this study or part of the laboratory collection, and their genotypes are listed in Table 5.1.

Genetic methods. (i) Transduction methods. The high-frequency generalized transducing mutant of bacteriophage P22 (HT105/1, int-201) (29) was used for all transductional crosses.

Transduction and subsequent purification was performed as previously described (30). (ii)

Mutant isolation. A pool of insertion mutants was created using the defective transposon

Tn10d(Tc) or Tn10d(Cm), as described previously (31). A P22 lysate was generated on the resulting mutant pool, and was used to transduce strain DM13513 (thiI897::MudJ(Kn)) or

DM10959 (thiI1205::Tn10d(Tc)) to tetracycline resistance or chloramphenicol resistance, respectively. Resulting antibiotic resistant colonies were screened for lack of growth on minimal cysteine/cystine and growth on minimal thiamine medium. The insertions were introduced into the parent strain and confirmed for appropriate phenotypes before determining the sequence adjacent to the insertion (32).

Molecular techniques. Plasmids were constructed using standard molecular techniques. DNA was amplified using Herculase (Agilent, Santa Clara, CA) or Q5 (New England Biolabs,

Ipswich, MA) DNA polymerase. Primers were purchased from Integrated DNA Technologies,

Coralville, IA. Plasmids were isolated using the Wizard Plus SV Miniprep kit (Promega,

Madison, WI), and PCR products were purified using the PCR purification kit (Qiagen, Venlo,

Limburg). Restriction endonucleases were purchased from New England Biolabs, Ipswich, MA, and ligase was purchased from Thermo Scientific, Waltham, MA. pFZY1 is a mini-F derivative

(averages 1-2 copies per cell) with a multiple-cloning site upstream of a promoterless galK9-

112

TABLE 5.1 Strains, plasmids and primers used in this study.

Strain number Genotype DM10959 thiI1205::Tn10d(Tc) DM13513 thiI897::MudJ(Kn) DM13559 thiI897::MudJ(Kn) ydjN81::Tn10d(Tc) DM13560 thiI897::MudJ(Kn) ydjN82::Tn10d(Tc) DM13868 ΔcysM::Kn ΔcysK::Cm DM14210 pDM1362 DM14211 pDM1363 DM14212 pDM1364 DM14237 thiI1205::Tn10d(Tc) ΔcdsH1::Cm DM14258 thiI1205::Tn10d(Tc) ΔcdsH1::Cm ΔaraCBAD::Kn pBAD24 thiI897::MudJ(Kn) ΔmetC2771 ΔSTM1557 ΔcysK::Cm ΔcysM::Kn DM14468 metB869::Tn10(Tc) Plasmid name Description pDM1363 pFZY1-cysPp pDM1362 pFZY1-ydjNp pDM1364 pFZY1-ydeDp pDM1371 pBAD24-cdsH pDM1375 pET14b-cdsH Primer name Sequence ydjNp KpnI for GGGTACCCAACGTCATACGCTGGT ydjNp BamHI rev CGGATCCCGCAATGTTCGCAATTAA cysPp KpnI for GGTACCTGATGGCGGCAGTAC cysPp BamHI rev CGGATCCCAGGGGTCTCTTTTTCAG ydeDp KpnI for GGGTACCTCGGGACGTTTTAATCTACCG ydeDp BamHI rev CGGATCCCAGTAACGCCAAAAAACCG cdsH NcoI for GAGACCATGGCCATGAGTAGCAATTGGGTTAA cdsH XbaI rev GAGATCTAGACTAGTCGCCGGTAAGTAATT cdsH NdeI for GAGACATATGATGAGTAGCAATTGGGTTAA cdsH XhoI rev GAGACTCGAGCTAGTCGCCGGTAAGTAATT

113 lacZYA reporter segment (33). The promoter regions of ydjN, cysP, and ydeD were cloned into pFZY1 using the KpnI and BamHI sites. cdsH was cloned into pBAD24 (34) using the NcoI and

XbaI sites for complementation and into pET14b (EMD Millipore, Millerca, MA) using the NdeI and XhoI sites for overproduction.

Nutritional analysis. Nutritional requirements were determined by liquid growth analyses and on solid media. For liquid growth analyses, strains were grown 6-12 h in 2 mL NB, harvested by centrifugation and resuspended in an equal volume of saline (0.15%). Five µL of cell suspension was used to inoculate 195 µL of the appropriate minimal media in each well of a 96-well microplate. Growth at 37°C with shaking at intensity level 2 was monitored using a microplate spectrophotometer (BioTek Instruments, Winooski, VT). When sulfur-containing compounds were included in the culture medium, at least two wells space was maintained between conditions due to volatility of those compounds.

β-galactosidase assays. β-galactosidase assays were performed with modifications of previously described methods (35, 36). Two µL of overnight culture grown in NB medium with ampicillin

(30 mg/L) was inoculated into 200 µL minimal glucose medium with ampicillin, with and without cystine. The cultures were grown in a BioTek microplate spectrophotometer until OD650

= 0.35-0.55, at which time 175-µL samples were removed to measure OD600. Twenty-µL samples of cells were then incubated with 80 µL permeabilization solution that contained Na2HPO4 (100 mM), KCl (20 mM), MgSO4 (2 mM), hexadecyltrimethylammonium bromide (1.6 mM), deoxycholic acid sodium salt (0.9 mM), and β-mercaptoethanol (77 mM). After 10 min incubation, 25 µL permabilized cell mixture was added to 150 µL substrate solution that contained Na2HPO4 (60 mM), NaH2PO4 (40 mM), o-nitrophenyl-β-D-galactoside (ONPG, 3.3 mM), and β-mercaptoethanol (39 mM). ONP product formation was monitored at A420 over time

114 in a SpectraMax 385 Plus microplate spectrophotometer (Molecular Devices, Sunnyvale, CA).

Miller units were calculated using the formula 1000*Vmax (ΔA420/min)/OD600/mL cell sample.

Purification of CdsH. CdsH was overproduced from pDM1375 in E. coli BL21AI (Life

Technologies, Carlsbad, CA). A 10-ml overnight culture in superbroth with ampicillin was used to inoculate 2 Fernbach flasks each containing 1.5-L of superbroth with ampicillin and grown at

37 °C until they reached OD650 of 0.4-0.7. Expression was induced by the addition of 0.02% arabinose and cultures were incubated at 22 °C with shaking for 18 h. Cells were harvested at 4

°C by centrifugation (15 minutes at 8,000 x g) and resuspended in Binding Buffer (50 mM

KPO4, pH 7.5, 100 mM NaCl). Lysozyme (1 mg/ml), phenylmethylsulfonyl fluoride (100 µg/ml) and DNase (25 µg/ml) were added to the cell suspension and incubated on ice for 1 h. Cells were mechanically lysed using a French pressure cell (5 passes at 10,342 kPa). The lysate was clarified by centrifugation (45 minutes at 48,000 x g), filtered through a 0.45-mm cellulose acetate membrane (Whatman, GE Healthcare Life Sciences, Piscataway, NJ) and loaded onto a column with 5 mL Ni-NTA Superflow and purified according to the manufacturer’s protocol

(Qiagen). Eluted protein was dialyzed into Dialysis Buffer 1 (50 mM KPO4 pH 7.5 100 mM

NaCl), then Dialysis Buffer 2 (50 mM KPO4 pH 7.5), then overnight in Freezing Buffer (50 mM

KPO4 pH 7.5 10% glycerol). Dialyzed protein was dropped into liquid nitrogen and frozen beads were stored at -80 °C. Purified CdsH was approximately 1.6 mg/ml as determined by bicinchinoic acid assay (Thermo Scientific) using bovine serum albumin as a standard.

CdsH activity assays. A typical reaction monitoring release of pyruvate at A230 included Tris buffer (50 mM, pH 8.6), PLP (25 µM), purified CdsH (0.005-5 µM) and cysteine, cystine or S- sulfocysteine (1 mM). Alternatively, formation of thiocysteine was measured based on detection of its free thiol with 5,5'-dithiobis-(2-nitrobenzoic acid) (DTNB) (37). When DTNB interacts

115

2- with a free thiol it releases 3-thio-6-nitrobenzoate (TNB ), which was monitored at A412 (the λmax of TNB2-). A typical reaction of this assay included potassium phosphate buffer (100 mM, pH

9.0) or Tris buffer (50 mM, pH 9.0), PLP (25 µM), DTNB (0.1 mM), purified CdsH (0.005-5

µM) and cystine or S-sulfocysteine (1 mM). Both assays were also tested at pH 6.5-10.5 using a mix of MES, HEPES and TAPS (50 mM each) at pH 6.5, 7.0, 7.5, 8.0 and 8.5, and a mix of

TAPS, CHES and CAPS (50 mM each) at pH 8.5, 9.0, 9.5, 10.0 and 10.5. In all cases, absorbance was monitored over time in a quartz 96-well plate in a SpectraMax microplate spectrophotometer (Molecular Devices, Sunnyvale, CA).

5.4 RESULTS AND DISCUSSION

The YdjN transporter is required for ThiI-independent thiamine biosynthesis. Mutational analysis was performed to identify genes required for ThiI-independent growth on minimal cysteine medium. A transposon library was introduced into a thiI mutant strain by transduction and resulting antibiotic-resistant colonies were screened for those that were unable to grow on minimal medium containing cysteine and retained growth on minimal medium containing thiamine. Approximately ~50,000 colonies were screened and two such mutants were found.

Both mutants contained an insertion in the ydjN gene. YdjN is homologous to the cystine transporter TcyP in Bacillus subtilis (38). In Escherichia coli, YdjN was reported to be the sole

S-sulfocysteine transporter and capable of transporting cystine but not cysteine (39). Cysteine is oxidized to the cysteine disulfide cystine (Cys-S-S-Cys) in the presence of atmospheric oxygen,

- and can be further oxidized to the thiosulfate S-sulfocysteine (Cys-S-SO3 ). The isolation of ydjN mutants and the previously characterized YdjN specificity suggested that cystine and S- sulfocysteine were the relevant components of the prepared cysteine stock that allowed growth.

116

Cystine and S-sulfocysteine support thiamine-independent growth of a thiI mutant. Cystine and S-sulfocysteine were added to the growth medium of a thiI mutant strain. In the presence of either compound, the thiI mutant strain displayed growth similar to that achieved with the addition of thiamine (Fig. 5.1A). In contrast, when the ydjN insertion was present in the thiI mutant strain, growth with S-sulfocysteine was abolished and growth with cystine was attenuated

(Fig. 5.1B). These data were consistent with the specificity profile reported for YdjN and suggested that in S. enterica this protein was the sole S-sulfocysteine transporter and one of the previously described cystine transport systems (CTS) (40). Attempts to clarify the role that cysteine had in allowing growth failed to produce consistent results. In total, the data were most consistent with oxidized products of cysteine, those transported by YdjN, being responsible for the thiamine independent growth of the thiI mutant. ydjN is a member of the cysteine regulon and encodes CTS-1. Three cystine transport systems

(CTS) have been described in S. enterica; only CTS-1 is a member of the cysteine regulon (40).

The effect of cysteine on the transcription of ydjN was determined. Two well-characterized promoters were used as controls: cysP encodes a thiosulfate transporter subunit and is a member of the cysteine regulon (41) and ydeD encodes an L-cysteine exporter that is not a member of the cysteine regulon (42). Like the cysPp control, expression from the ydjNp was repressed when the growth medium included cystine, which represses expression of cysteine regulon genes (Table

5.2) (43). These data supported the conclusion that ydjN was a member of the cysteine regulon and encoded CTS-1.

CdsH is required for ThiI-independent thiamine biosynthesis. Considering the chemistry mediated by ThiI, it was feasible that a non-protein small molecule persulfide (e.g. thiocysteine

(Cys-S-S-H)) could replace ThiI as the sulfur donor to the activated ThiS terminal glycine. This

117

A 0.5 B 0.5 650 650 OD OD

0 10 20 0 10 20 Time (h) Time (h)

FIGURE 5.1 A ydjN mutation impacts ThiI-independent growth. A thiI mutant strain (A) or thiI ydjN mutant strain (B) was grown in minimal medium with glucose (11 mM) and NaNO3 (25 mM) as an electron acceptor. The cultures were grown with the following additions: none (circles); 100 µM S-sulfocysteine (inverted triangles); 100 µM cystine (triangles); 100 nM thiamine (squares). Growth was monitored by optical density at 650 nm without shaking at 37°C to help prevent oxidation of cystine. Data are representative growth curves from experiments repeated on two separate days in biological triplicate.

118

TABLE 5.2 Expression from the ydjNp is reduced when cystine is present in the medium. β- galactosidase assays were performed on cultures in mid-log phase grown in minimal glucose medium with and without cystine (100 µM) provided. The data are reported in Miller Units -1 -1 -1 (1000 X A420 min OD600 ml ) and represent averages and standard deviations from three independent cultures.

Promoter - Cystine + Cystine

cysPp 1470 ± 170 311 ± 28

ydeDp 370 ± 60 339 ± 32

ydjNp 1320 ± 130 353 ± 11

119 scenario would be consistent with the requirement for ThiF in ThiI-independent thiamine biosynthesis (18). In the simplest form of this scenario, cystine (and/or S-sulfocysteine) would be converted non-enzymatically to thiocysteine by pyridoxal-5-phosphate (PLP) and metal ions (44) or by PLP-dependent cystine lyases. S. enterica encodes several enzymes with cystine lyase activity, including the γ-cystathionase MetB (45), and the β-cystathionases MetC (37) and

STM1557 (MalY) (46). A strain lacking all of these activities was constructed and tested for

ThiI-independent growth in the presence of cysteine. The thiI metC STM1557 cysM cysK metB mutant strain grew in the absence of thiamine if cystine was provided (data not shown) ruling out a role for the cystine lyases.

However, neither cystine nor S-sulfocysteine were able to support ThiI-independent thiamine synthesis in a strain lacking the cdsH gene (Fig. 5.2). CdsH is the dominant cysteine desulfhydrase in S. enterica and converts cysteine to pyruvate, sulfide and ammonia (47-49).

Despite the canonical activity of CdsH, it was formally possible that CdsH was acting on an alternative substrate, (i.e. cystine or S-sulfocysteine) and generated a persulfide needed for thiamine synthesis. CdsH was purified and analyzed for this potential activity. When provided at

1 mM, neither cystine nor S-sulfocysteine served as substrate for CdsH. Despite demonstrable desulfhydrase activity with cysteine, neither alternative substrates generated detectable pyruvate

(increase at A230) or any thiol that could be derivitized by 5,5'-dithiobis-(2-nitrobenzoic acid)

(DTNB) (increase at A412). In addition, no evidence of a persulfide was detected when the assay mixture contained buffer at pH 6.5-10.5.

Model for persulfide formation in the thiI mutant strain. Considering the data in total, a model for ThiI-independent thiamine can be proposed. This model must account for several results herein and previous findings. First, it must consider the fact that cystine or S-

120

A B 1 1 650 650 OD OD

0.1 0.1

0 5 10 0 5 10 Time (h) Time (h)

FIGURE 5.2 CdsH is required for cystine or S-sulfocysteine-stimulated growth. A cdsH thiI mutant strain with empty vector (pBAD24, open symbols) or pCdsH (filled symbols) was grown in minimal medium with glucose (11 mM), 0.02% arabinose and 100 µM cystine (Panel A) or 100 µM S-sulfocysteine (Panel B), with thiamine (diamonds) or without thiamine (circles). Growth was monitored by optical density at 650 nm with shaking at 37°C. Data represent the average and standard deviation from three independent cultures.

121 sulfocysteine, but not cysteine, allows thiamine synthesis. This result is informative since both cystine and S-sulfocysteine are expected to be reduced to cysteine in the cellular environment via glutathione or thioredoxin dependent mechanisms (50, 51). Secondly, the model must include a role for CdsH, and finally, a persulfide is implicated by the fact that ThiFS are required for the thiamine synthesis (18).

The model depicted in Figure 5.3 is consistent with each of these results and includes the following key features. Cystine and S-sulfocysteine are transported by YdjN, after which a significant proportion of each is reduced to cysteine (Fig. 3A). The cysteine serves as a substrate for CdsH, generating products amoninum, pyruvate and the important sulfide (Fig. 3B). The sulfide then attacks the sulfane sulfur in the remaining cystine or S-sulfocysteine to generate the small molecule persulfide, thiocysteine. Sulfide can generate various persulfides by interacting with different disulfides in basic conditions, including thiocysteine from cystine (52), a lipoate persulfide from oxidized lipoate (53), protein persulfides from disulfide bonds (54) and glutathione persulfide from oxidized glutathione (55). After sulfide attack, the resulting persulfide (e.g. thiocysteine) then replaces the ThiI persulfide in donating sulfur to the ThiF-ThiS acyldisulfide (Fig. 3C).

Although the relevant sulfide species is not known, it is depicted as the sulfide ion for the following reasons: sulfide-dependent persulfide formation requires basic conditions, suggesting the relevance of the dissociated HS- or S2- species. Second, CdsH could not be replaced by addition of Na2S to the growth medium (data not shown), which would be present as H2S and

HS- at physiological pH of 7.4 (56). This general model also allows for the potential that sulfide interacts with mixed disulfide intermediates in cystine and S-sulfocysteine metabolism, Cys-S-S-

Glutathione or Cys-S-S-Thioredoxin (51, 57) to produce a persulfide.

122

FIGURE 5.3 Model for ThiI-independent THZ-P biosynthesis. In this model, YdjN transports cystine and/or S-sulfocysteine into the cell, where they are then metabolized to cysteine. CdsH converts cysteine to pyruvate, ammonia and sulfide. Sulfide could then attack the sulfur atom in a disulfide (R1-S-S-R2), releasing a persulfide (R1-S-S-H) and a thiol (R2-S-H). This disulfide could include cystine, S-sulfocysteine, oxidized glutathione or a mixed disulfide. The persulfide could then replace ThiI in sulfur donation to ThiFS, resulting in a ThiFS acyldisulfide and releasing a second thiol. This model is based on the assumption that a persulfide functional group is required to replace sulfur donation by the ThiI persulfide. Abbreviations: Ox, oxidation; GSH, glutathione; GSSG, oxidized glutathione; Trx, thioredoxin; Trx-S-S-Trx, oxidized thioredoxin.

123

ox A Cystine S-sulfocysteine

YdjN inside cell

-OOC Cystine S-sulfocysteine -OOC H2N S S NH2 H2N SO - - S 3 GSH COO

GSSG GSH/Trx-SH

GSSG/Trx-SS-Trx Cysteine 2- - SO3 OOC NH2

SH B Cysteine

CdsH H 2- + NH4 + S R1 S S R2 R2 SH + Pyruvate

R1 Persulfide S C S H O ATP PPi O ThiS ThiS O O AMP ThiF ThiF Cys184 SH Cys184 SH

R1 SH R H 1 S O ThiS S ThiS O S ThiF ThiF Cys184 S Cys184 S H

124

Conclusions. Use of a small molecule persulfide for sulfur trafficking could be representative of thiazole biosynthesis in some organisms that lack the ThiI rhodanese domain. For example, in the marine genus Synechocystis the cystine lyase C-DES uses cystine to generate thiocysteine, which may provide persulfide sulfur in the cell (58). Although SufS is thought to be the major source of persulfide sulfur in the Synechocystis, C-DES is important for growth under dim light conditions (59). Synechocystis sp. PCC 6803 encodes all thiamine biosynthetic enzymes except

ThiI and IscS and does not require thiamine supplementation in the medium (60). This suggests that thiocysteine formation from cystine could provide activated persulfide sulfur to thiamine biosynthesis during growth in dim light conditions when C-DES is important for growth.

Additional variations on this theme could include small molecule persulfides serving as sulfur donors to 4-thiouridine, iron-sulfur clusters and other sulfur-containing nucleosides and cofactors.

5.5 ACKNOWLEDGEMENTS

We thank William Whitman for helpful discussion of this work. The National Institutes of Health grant GM47296 to DMD supported this work and LDP was supported by NSF through

Graduate Research Fellowship grant DGE-0718123.

Man Him (Sammy) Leung isolated and identified the ydjN mutants and initially characterized the cystine and S-sulfocysteine rescue. I performed all of the remaining experiments described in this chapter.

5.6 REFERENCES

2. Raschke M, Burkle L, Muller N, Nunes-Nesi A, Fernie AR, Arigoni D, Amrhein N, Fitzpatrick TB. 2007. Vitamin B1 biosynthesis in plants requires the essential iron sulfur cluster protein, THIC. Proc. Natl. Acad. Sci. U. S. A. 104:19637-19642.

125

3. Reddick JJ, Kinsland C, Nicewonger R, Christian T, Downs DM, Winkler ME, Begley TP. 1998. Overexpression, purification and characterization of two pyrimidine kinases involved in the biosynthesis of thiamin: 4-amino-5-hydroxymethyl-2-methylpyrimidine kinase and 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate kinase. Tetrahedron 54:15983-15991.

4. Petersen LA, Downs DM. 1997. Identification and characterization of an operon in Salmonella typhimurium involved in thiamine biosynthesis. J. Bacteriol. 179:4894-4900.

5. Vander Horn PB, Backstrom AD, Stewart V, Begley TP. 1993. Structural genes for thiamine biosynthetic enzymes (thiCEFGH) in Escherichia coli K-12. J. Bacteriol. 175:982-992.

6. Webb E, Claas K, Downs DM. 1997. Characterization of thiI, a new gene involved in thiazole biosynthesis in Salmonella typhimurium. J. Bacteriol. 179:4399-4402.

7. Lauhon CT, Kambampati R. 2000. The iscS gene in Escherichia coli is required for the biosynthesis of 4-thiouridine, thiamin and NAD. J. Biol. Chem. 275:20096-20103.

8. Settembre EC, Dorrestein PC, Park JH, Augustine AM, Begley TP, Ealick SE. 2003. Structural and mechanistic studies on ThiO, a glycine oxidase essential for thiamin biosynthesis in Bacillus subtilis. Biochemistry 42:2971-2981.

9. Rajakovich LJ, Tomlinson J, Dos Santos PC. 2012. Functional analysis of Bacillus subtilis genes involved in the biosynthesis of 4-thiouridine in tRNA. J. Bacteriol. 194:4933-4940.

10. Backstrom AD, Mcmordie RAS, Begley TP. 1995. Biosynthesis of thiamin I: the function of the thiE gene product. J. Am. Chem. Soc. 117:2351-2352.

11. Webb E, Downs D. 1997. Characterization of thiL, encoding thiamin-monophosphate kinase, in Salmonella typhimurium. J. Biol. Chem. 272:15702-15707.

12. Kluger R. 1987. Thiamin diphosphate - a mechanistic update on enzymatic and nonenzymic catalysis of decarboxylation. Chem Rev 87:863-876.

13. Lipsett MN. 1978. Enzymes producing 4-thiouridine in Escherichia coli tRNA: approximate chromosomal locations of the genes and enzyme activities in a 4- thiouridine-deficient mutant. J. Bacteriol. 135:993-997.

14. Mueller EG, Buck CJ, Palenchar PM, Barnhart LE, Paulson JL. 1998. Identification of a gene involved in the generation of 4-thiouridine in tRNA. Nucleic Acids Res. 26:2606- 2610.

15. Mueller EG, Palenchar PM. 1999. Using genomic information to investigate the function of ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis. Prot. Sci. 8:2424-2427.

126

16. Aravind L, Koonin EV. 2001. THUMP--a predicted RNA-binding domain shared by 4- thiouridine, pseudouridine synthases and RNA methylases. Trends Biochem. Sci. 26:215- 217.

17. Waterman DG, Ortiz-Lombardia M, Fogg MJ, Koonin EV, Antson AA. 2006. Crystal structure of Bacillus anthracis ThiI, a tRNA-modifying enzyme containing the predicted RNA-binding THUMP domain. J. Mol. Biol. 356:97-110.

18. Martinez-Gomez NC, Palmer LD, Vivas E, Roach PL, Downs DM. 2011. The rhodanese domain of ThiI is both necessary and sufficient for synthesis of the thiazole moiety of thiamine in Salmonella enterica. J. Bacteriol. 193:4582-4587.

19. You D, Xu T, Yao F, Zhou X, Deng Z. 2008. Direct evidence that ThiI is an ATP pyrophosphatase for the adenylation of uridine in 4-thiouridine biosynthesis. Chem. Biol. Chem. 9:1879-1882.

20. Bordo D, Bork P. 2002. The rhodanese/Cdc25 phosphatase superfamily. Sequence- structure-function relations. EMBO reports 3:741-746.

21. Liu Y, Zhu X, Nakamura A, Orlando R, Soll D, Whitman WB. 2012. Biosynthesis of 4- thiouridine in tRNA in the methanogenic archaeon Methanococcus maripaludis. J. Biol. Chem. 287:36683-36692.

22. Taylor SV, Kelleher NL, Kinsland C, Chiu HJ, Costello CK, Backstrom AD, McLafferty FW, Begley TP. 1998. Thiamin biosynthesis in Escherichia coli - Identification of ThiS thiocarboxylate as the immediate sulfur donor in the thiazole formation. J. Biol. Chem. 273:16555-16560.

23. Xi J, Ge Y, Kinsland C, McLafferty FW, Begley TP. 2001. Biosynthesis of the thiazole moiety of thiamin in Escherichia coli: identification of an acyldisulfide-linked protein-- protein conjugate that is functionally analogous to the ubiquitin/E1 complex. Proc. Natl. Acad. Sci. U. S. A. 98:8513-8518.

24. Kambampati R, Lauhon CT. 2000. Evidence for the transfer of sulfane sulfur from IscS to ThiI during the in vitro biosynthesis of 4-thiouridine in Escherichia coli tRNA. J. Biol. Chem. 275:10727-10730.

25. Palenchar PM, Buck CJ, Cheng H, Larson TJ, Mueller EG. 2000. Evidence that ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis, may be a sulfurtransferase that proceeds through a persulfide intermediate. J. Biol. Chem. 275:8283-8286.

26. Mueller EG. 2006. Trafficking in persulfides: delivering sulfur in biosynthetic pathways. Nat. Chem. Biol. 2:185-194.

27. Vogel HJ, Bonner DM. 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218:97-106.

127

28. Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260-296.

29. Schmieger H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-88.

30. Downs DM, Petersen L. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.

31. Petersen L, Enos-Berlage J, Downs DM. 1996. Genetic analysis of metabolic crosstalk and its impact on thiamine synthesis in Salmonella typhimurium. Genetics 143:37-44.

32. Caetano-Anolles G. 1993. Amplifying DNA with arbitrary oligonucleotide primers. PCR methods and applications 3:85-94.

33. Koop AH, Hartley ME, Bourgeois S. 1987. A low-copy-number vector utilizing β- galactosidase for the analysis of gene control elements. Gene 52:245-256.

34. Guzman L-M, Belin D, Carson MJ, Beckwith J. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.

35. Griffith KL, Wolf RE. 2002. Measuring β-galactosidase activity in bacteria: cell growth, permeabilization, and enzyme assays in 96-well arrays. Biochim. Biophys. Acta 290:397- 402.

36. Zhang X, Bremer H. 1995. Control of the Escherichia coli rrnB P1 promoter strength by ppGpp. J. Biol. Chem. 270:11181-11189.

37. Dwivedi CM, Ragin RC, Uren JR. 1982. Cloning, purification, and characterization of β- cystathionase from Escherichia coli. Biochemistry 21:3064-3069.

38. Burguiere P, Auger S, Hullo MF, Danchin A, Martin-Verstraete I. 2004. Three different systems participate in L-cystine uptake in Bacillus subtilis. J. Bacteriol. 186:4875-4884.

39. Nonaka K. 2010. L-cysteine-producing bacterium and a method for producing L- cysteine. Pub. No. US 2010/0233765.

40. Baptist EW, Kredich NM. 1977. Regulation of L-cystine transport in Salmonella typhimurium. J. Bacteriol. 131:111-118.

41. Hryniewicz MM, Kredich NM. 1991. The cysP promoter of Salmonella typhimurium: characterization of two binding sites for CysB protein, studies of in vivo transcription initiation, and demonstration of the anti-inducer effects of thiosulfate. J. Bacteriol. 173:5876-5886.

128

42. Ohtsu I, Wiriyathanawudhiwong N, Morigasaki S, Nakatani T, Kadokura H, Takagi H. 2010. The L-cysteine/L-cystine shuttle system provides reducing equivalents to the periplasm in Escherichia coli. J. Biol. Chem. 285:17479-17487.

43. Kredich NM. 1971. Regulation of L-cysteine biosynthesis in Salmonella typhimurium. I. Effects of growth of varying sulfur sources and O-acetyl-L-serine on gene expression. J. Biol. Chem. 246:3474-3484.

44. Cavallini D, De Marco C, Mondovi B. 1960. Cleavage of cystine by a pyridoxal model. Arch. Biochem. Biophys. 87:281-288.

45. Delavier-Klutchko C, Flavin M. 1965. Role of a bacterial cystathionine-β-cleavage enzyme in disulfide decomposition. Biochim. Biophys. Acta 99:375-377.

46. Zdych E, Peist R, Reidl J, Boos W. 1995. MalY of Escherichia coli is an enzyme with the activity of a β-C-S lyase (cystathionase). J. Bacteriol. 177:5035-5039.

47. Oguri T, Schneider B, Reitzer L. 2012. Cysteine catabolism and cysteine desulfhydrase (CdsH/STM0458) in Salmonella enterica serovar Typhimurium. J. Bacteriol. 194:4366- 4376.

48. Kredich NM, Keenan BS, Foote LJ. 1972. The purification and subunit structure of cysteine desulfhydrase from Salmonella typhimurium. J. Biol. Chem. 247:7157-7162.

49. Kredich NM, Foote LJ, Keenan BS. 1973. The stoichiometry and kinetcis of the inducible cysteine desulfhydrase from Salmonella typhimurium. J. Biol. Chem. 248:6187- 6196.

50. Park JH, Dorrestein PC, Zhai H, Kinsland C, McLafferty FW, Begley TP. 2003. Biosynthesis of the thiazole moiety of thiamin pyrophosphate (vitamin B1). Biochemistry 42:12430-12438.

51. Funane K, Iwahashi H, Nakamura T. 1987. Metabolism of S-sulfocysteine in Salmonella typhimurium. Role of thioredoxin in the reduction of S-sulfocysteine. Agric. Biol. Chem. 51:1247-1256.

52. Rao GS, Gorin G. 1959. Reaction of cystine with sodium sulfide in sodium hydroxide solution. J. Org. Chem. 24:749-753.

53. Villarejo M, Westley J. 1963. Mechanism of rhodanese catalysis of thiosulfate-lipoate oxidation-reduction. J. Biol. Chem. 238:4016-4020.

54. Cavallini D, Federici G, Barboni E. 1970. Interaction of proteins with sulfide. Eur. J. Biochem. 14:169-174.

55. Francoleon NE, Carrington SJ, Fukuto JM. 2011. The reaction of H2S with oxidized thiols: generation of persulfides and implications to H2S biology. Arch. Biochem. Biophys. 516:146-153.

129

56. Reiffenstein RJ, Hulbert WC, Roth SH. 1992. Toxicology of hydrogen sulfide. Annu. Rev. Pharmacol. Toxicol. 32:109-134.

57. Woodin TS, Segel IH. 1968. Glutathione reductase-dependent metabolism of cysteine-S- sulfate by Penicillium chrysogenum. Biochim. Biophys. Acta 167:78-88.

58. Lang T, Kessler D. 1999. Evidence for cysteine persulfide as a reaction product of L- cyst(e)ine C-S-lyase (C-DES) from Synechocystis. J. Biol. Chem. 274:189-195.

59. Kessler D. 2006. Enzymatic activation of sulfur for incorporation into biomolecules in prokaryotes. FEMS Microbiol. Rev. 30:825-840.

60. Stanier RY, Kunisawa R, Mandel M, Cohen-Bazire G. 1971. Purification and properties of unicellular blue-green algae (order Chroococcales). Bacteriol. Rev. 35:171-205.

130

CHAPTER 6

SACCHAROMYCES CEREVISIAE THIAMINE PYRIMIDINE SYNTHASE FUNCTIONS

CONDITIONALLY DURING HETEROLOGOUS EXPRESSION

IN SALMONELLA ENTERICA1

1 Palmer L.D., D.M. Downs. To be submitted to Journal of Bacteriology.

131

6.1 ABSTRACT

Thiamine pyrophosphate is a required cofactor for all forms of life. The pyrimidine moiety of thiamine, 2-methyl-4-amino-5-hydroxymethylpyrimidine phosphate (HMP-P), is synthesized by different mechanisms in bacteria and plants compared to fungi. In this study,

Salmonella enterica was used to probe requirements for activity of the yeast HMP-P synthase,

Thi5p. Thi5p synthesizes HMP-P from histidine and pyridoxal-5-phosphate and was reported to use its His66 as the histidine substrate, suggesting it was a single turnover enzyme. Expression of

THI5 could only complement a S. enterica HMP-P auxotroph during growth on certain carbon sources; production of Thi5p did not support growth on minimal glucose medium. To investigate the metabolic factors affecting Thi5p activity, suppressors were isolated that allowed the S. enterica THI5 strain to grow on glucose medium. Further genetic analysis suggested that reduction of glucose transport/phosphorylation allowed Thi5p-dependent growth on minimal glucose medium, but that glycolytic intermediates were not inhibiting Thi5p directly.

6.2 INTRODUCTION

Thiamine pyrophosphate (TPP) is a cofactor for many central metabolic enzymes, and is required at low levels by all organisms. Humans require dietary intake of thiamine, which is biosynthesized by a variety of plants, bacteria, and fungi. TPP is composed of two independently synthesized moieties, 5-(2-Hydroxyethyl)-4-methylthiazole phosphate (THZ-P) and 2-methyl-4- amino-5-hydroxymethylpyrimidine phosphate (HMP-P). In bacteria and plants, the first steps of the HMP-P biosynthesis pathway are shared with purine biosynthesis (Fig. 6.1) (1, 2). In these organisms, the radical S-adenosylmethionine enzyme ThiC catalyzes an intramolecular rearrangement of the purine intermediate 5-aminoimidazole ribotide (AIR) to the pyrimidine

HMP-P (3, 4). 132

Purines COOH ThiC

H2N

H2N HN N N Thi5p Histidine PO N

O HMP-P THZ-P HO

PO N TPP PLP FIGURE 6.1 Schematic representation of Thi5p-dependent thiamine biosynthesis in S. enterica. The genes encoding the relevant enzymes are shown. Abbreviations: THZ-P, 5-(2- Hydroxyethyl)-4-methylthiazole phosphate;HMP-P, 2-methyl-4-amino-5- hydroxymethylpyrimidine phosphate. 133

In contrast, fungi do not contain a ThiC homolog; rather, the Thi5p enzyme family synthesizes HMP-P under aerobic conditions. In vivo labeling in yeast implicated histidine and pyridoxine as precursors to HMP-P biosynthesis under aerobic conditions (5-7). There are four members of THI5 gene family in Saccharomyces cerevisiae (THI5 (YFL058w), THI11

(YJR156c), THI12 (YNL332w) and THI13 (YDL244w)), but other species have between zero and five gene copies (8). Genetic analysis in S. cerevisiae demonstrated that the four enzymes are functionally redundant, as only the quadruple mutant displayed thiamine auxotrophy under aerobic conditions (8). A less efficient HMP-P biosynthetic pathway operates under anaerobic conditions and does not rely on pyrodoxine or Thi5p family enzymes (8, 9).

The labeling pattern for aerobic HMP-P synthesis in S. cerevisiae suggested the involvement of unique chemistry, but in vitro reconstitution of the reaction lagged behind labeling studies. Lai, et al., reported reconstituted Candida albicans Thi5p activity in vitro and found that, paradoxically, the protein was oxygen sensitive but required oxygen, iron and PLP for HMP-P synthesis. These authors found that added histidine was not required for Thi5p activity and their data that suggested His66 served as the histidine substrate, implying Thi5p is a single-turnover enzyme (10). The importance of His66 for Thi5p activity was corroborated by

Coquille, et al., who reported His66 was required for Thi5 activity in vivo in S. cerevisiae (11).

The finding that Thi5p is a single-turnover enzyme raises questions about differences in physiology between fungi and bacteria/plants that select for maintenance of Thi5p over ThiC, particularly in light of evidence of horizontal gene transfer from bacteria to S. cerevisiae (12).

Previous physiological studies have led to successful reconstitution or improvement of in vitro activity for enzymes that are difficult to study. For example, the ThiC bacterial HMP-P synthase activity in vivo was linked to methionine and iron sulfur cluster metabolism in 134

Salmonella enterica (2, 13, 14), leading to its subsequent reconstitution of activity and identification as a Radical SAM enzyme (3, 4). Similarly, although many reports described biotin synthase BioB as a single-turnover enzyme in vitro, in vivo work in E. coli showed that BioB was capable of multiple turnovers and inhibited by its product 5’-deoxyadenosine (15, 16). These findings informed later studies demonstrating that BioB was capable of multiple turnovers in vitro when product inhibition was alleviated (17).

Here we describe characterization of S. cerevisiae Thi5p activity in the model organism

S. enterica. This study was initiated with two goals: to begin to describe metabolic network differences between two well-studied organisms of different domains; and to better understand

Thi5p by studying its interaction with S. enterica physiology. Metabolic networks are thought to be composed of conserved metabolic modules (18), and general network organizing principles are conserved broadly (19). However, it remains unclear the degree to which metabolic crosstalk or integration is conserved. Because thiamine biosynthesis in S. enterica is an established system for metabolic dissection (reviewed in (20)), analysis of Thi5p activity in S. enterica could uncover metabolic differences between S. enterica and S. cerevisiae.

6.3 MATERIALS AND METHODS

Strains, media and chemicals. Minimal medium was No-carbon E medium (NCE) (21)

supplemented with MgSO4 (1 mM), trace minerals (0.1X) (22) and carbon sources based on 66 mM available carbon units with the following exceptions: pyruvate (50 mM, prepared from powder as needed), acetate (as noted), malate (40 mM), succinate (20 mM) fumurate (50 mM), serine (10 mM). When noted, additions were at the following concentrations: thiamine, 100 nM; histidine, 75 μM; pyridoxine vitamers, 100 μM; casamino acids, 1% w/v; α-methylglucoside,

0.5% w/v; 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (XGal), 20 mg/L; isopropyl β-D- 135

1-thiogalactopyranoside (IPTG), 1 mM. Rich media was Difco nutrient broth (NB; 8 g/L) with

NaCl (5 g/L), lysogeny broth (LB) or superbroth (SB; tryptone (32 g/L), yeast extract (20 g/L),

NaCl (5 g/L) with NaOH (0.05 N)). Solid media contained 1.5% agar. Antibiotics were added at the following concentrations in rich and minimal media, respectively, unless otherwise noted: chloramphenicol (Cm), 20 mg/L, 5 mg/L; ampicillin (Ap), 150 mg/L, 30 mg/L; tetracycline (Tc),

20 mg/L, 10 mg/L; kanamycin (Kn), 50 mg/L, 12.5 mg/L. All chemicals were purchased from

Sigma-Aldrich, St Louis, MO. The strains used in this study were derivatives of S. enterica strain

LT2 and were generated for this study or part of the laboratory collection, and their genotypes are listed in Table 6.1.

Genetic methods. The high-frequency generalized transducing mutant of bacteriophage P22

(HT105/1, int-201) (23) was used for all transductional crosses. Transduction and subsequent purification was performed as previously described (24).

Molecular techniques. Plasmids were constructed using standard molecular techniques. DNA was amplified using Herculase (Agilent, Santa Clara, CA) or Q5 (New England Biolabs,

Ipswich, MA) DNA polymerase. Primers were purchased from Integrated DNA Technologies,

Coralville, IA. Plasmids were isolated using the Wizard Plus SV Miniprep kit (Promega,

Madison, WI), and PCR products were purified using the PCR purification kit (Qiagen, Venlo,

Limburg). Restriction endonucleases were purchased from New England Biolabs, Ipswich, MA, and ligase was purchased from Thermo Scientific, Waltham, MA.

The plasmids and primers are listed in Table 6.2. The THI5 yeast GST-tagged plasmid from Thermo Open Biosystems (Huntsville, AL) was isolated using the modified Qiagen miniprep protocol for yeast and used as template for amplification of THI5. Later analysis revealed the clone ID YDR155C plasmid included a mixed population of wild-type and variant THI5 136

TABLE 6.1 Bacterial strains.

Strain Genotype Number DM13623 ΔthiC1225 ΔaraCBAD pDM1336 DM13631 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc)a sgrR1 pDM1336 DM13632 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) pDM1336 DM13836 ΔthiC1225 ΔaraCBAD yfeA85::Tn10d(Tc) pDM1336 DM13837 ΔthiC1225 ΔaraCBAD yfeA85::Tn10d(Tc) ptsI611 pDM1336 DM14419 ΔthiC1225 ΔaraCBAD pDM1381 DM14422 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1381 DM14426 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) sgrR1 pDM1381 DM14427 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) pDM1381 DM14428 ΔthiC1225 ΔaraCBAD yfeA85::Tn10d(Tc) pDM1381 DM14429 ΔthiC1225 ΔaraCBAD yfeA85::Tn10d(Tc) ptsI611 pDM1381 DM14430 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d sgrR2 pDM1381 DM14431 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) pDM1381 DM14442 ΔthiC1225 ΔaraCBAD ptsI421::Tn10(Tc) pDM1381 DM14443 ΔthiC1225 ΔaraCBAD pgi::Tn5(Kn)b pDM1381 DM14463 ΔthiC1225 ΔaraCBAD zwf21::Tn10d(Tc) pDM1381 ΔthiC1225 ΔaraCBAD pgi::Tn5(Kn) zwf21::Tn10d(Tc) DM14464 pDM1381 DM14488 ΔthiC1225 ΔaraCBAD pDM1381 pBAD24 DM14489 ΔthiC1225 ΔaraCBAD pDM1381 pDM1398 DM14490 ΔthiC1225 ΔaraCBAD pDM1381 pDM1399 DM14491 ΔthiC1225 ΔaraCBAD pDM1381 pDM1400 DM14509 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1381 pBAD24 DM14510 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1381 pDM1398 DM14511 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1381 pDM1399 DM14412 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1381 pDM1400 DM14517 ΔthiC1225 ΔaraCBAD ptsG4152::Tn10(Tc) pDM1381 DM14531 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) pDM1402 DM14532 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) sgrR1 pDM1402 DM14533 ΔthiC1225 ΔaraCBAD zxx10167::Tn10d(Tc) sgrR2 pDM1402 DM14534 ΔthiC1225 ΔaraCBAD ΔsgrR3::Kn pDM1402 DM14539 ΔthiC1225 ΔaraCBAD crr-307::Tn10d(Tc) pDM1381 DM14540 ΔthiC1225 ΔaraCBAD ΔpfkA::Kn pDM1381 aTn10d(Tc) refers to the transposition-defective mini-Tn10(Tn10Δ16Δ17 tetR) (25)

137

TABLE 6.2 Plasmid and primers.

Plasmid Name Description pDM1336 pSU18-THI5 pDM1381 pSU18-THI5 pDM1361 pBAD18S-THI5 pDM1398 pBAD24-sgrR pDM1399 pBAD24-sgrR1 pDM1400 pBAD24-sgrR2 pDM1402 pFZY1-sgrSp Primer Name Sequence Sc THI5 for SacI pBAD18 AAGAGCTCTGATGTCTACAGACAAGATC Sc THI5 Rev HindIII TTTAAGCTTTTAAGCTGGAAGAGCCAATC sgrSp_KpnI_F ATGGTACC CATAAAAGGGGAACTC sgrSp_BamHI_R ATAGGATCCCGAAAGATATTATTGGC S. cerevisiae Thi5 for 5' CATATGATGTCTACAGACAAGATCAC S. cerevisiae Thi5 rev 3' CTCGAGTTAAGCTGGAAGAGCCAATC CATTCCAAAGTTCAGAGGTAGTCATGATTAAGAAAA pfkA_wanner_for TCGGGTGTAGGCTGGAGCTGCTTC CCATCAGGCGCGCAAAAACAATCAGTACAGTTTTTT pfkA_wanner_rev CGCGCATATGAATATCCTCCTTAG ATCAGCCTAACAGGAGGTAACGATGGTACGTATCTA pfkB_wanner_for TACGGTGTAGGCTGGAGCTGCTTC CTCGGCGAGGGGGAAACGATTATTGCGCGGAAAGA pfkB_wanner_rev TAGGCCATATGAATATCCTCCTTAG TTTTCATCGGAGTTCCCCTTTTATGCCCTCAGGTCGC yabN (sgrR) wanner for CTGGTGTAGGCTGGAGCTGCTTC CAGCAATCAAGAGCTGGCGTTAAGGATCTGGCGGCG yabN (sgrR) wanner rev CAAACATATGAATATCCTCCTTAG yabN(sgrR)_NcoI_for ATATCCATGGCAATGCCCTAGGTCGC yabN(sgrR)_XbaI_rev GCGATCTAGATTAAGGATCTGGCGGCGC

138

(cloned in pDM1336) with the following substitutions: M37I, A138V, G152D. We did not detect any functional differences between the variant THI5 or wild-type THI5. pSU18 is a pACYC184 derivative compatible with pBR322 that uses the lac promoter and encodes CmR

(26). To construct pDM1336 and pDM1381, THI5 was amplified using primers S. cerevisiae

Thi5 for 5' and S. cerevisiae Thi5 rev 3'; the resulting PCR products were cleaned up and blunt- end ligated into pSU18 digested with SmaI.

pBAD vectors are pBR322 derivatives that use the ara promoter and encode AmpS; pBAD18S includes the N-terminal sequence of AraB Met-Ala-Ile-Ala-Gly prior to the multiple cloning site (27). THI5 was cloned into SacI/HindIII sites in pBAD18S to construct pDM1361. sgrR alleles were cloned into NcoI/XbaI sites in pBAD24 to construct pDM1398-1400. pFZY1 is a mini-F derivative (averages 1-2 copies per cell) with a multiple-cloning site upstream of a promoterless galK9-lacZYA reporter segment (28). The promoter region of sgrS was cloned into pFZY1 using the KpnI/BamHI sites to construct pDM1402.

Isolation of thiC pTHI5 derivatives allowing Thi5p-dependent growth in minimal glucose medium. Ten independent cultures of DM13623 in NB Cm were incubated overnight with shaking at 37°C. A 100-µL sample (~ 1 X 108 cells) was spread on a minimal glucose medium plate and 5 µL diethyl sulfate (DES) was spotted in the center of the plate as indicated.

Suppressor frequency was analyzed after incubation at 37°C for 2 days. Six colonies per plate were streaked for individual colonies on selective medium (minimal glucose), then non-selective medium (NB Cm), before patching to NB Cm and printing to minimal glucose for confirmation of selected phenotype.

The resultant strains were separated into classes based on growth in minimal glucose medium. For each class, a transposon (Tn10d(Tc)) genetically linked to the causative mutation 139 was isolated by standard genetic techniques and used to reconstruct the mutant for phenotypic confirmation. The chromosomal location of relevant insertions was determined by sequencing using a PCR-based protocol (29). A DNA product was amplified with degenerate primers and primers derived from the Tn10d(Tc) insertion sequence and sequenced. The genome of the reconstructed strains was sequenced to identify the causative mutation (see below).

Alternatively, putative loci were PCR amplified and sequenced by GeneWiz (South Plainfield,

NJ). The strains were reconstructed in the DM14419 background.

Genome sequencing. Whole-genome sequencing was used to identify the causative suppressor mutations in DM13631 and DM13836. High molecular-weight genomic DNA was isolated using a phenol-chloroform extraction. A 1-ml culture was grown to full density in superbroth, harvested by centrifugation and resuspended in buffer (0.1 M Tris pH 8, 0.15 M NaCl and 0.1 M

EDTA). Lysozyme (0.5 mg) was added, and the sample was incubated at 37 °C for 10 minutes. Proteinase K (1 mg) and SDS (1 %) were added and incubation continued at 37 °C for

30 minutes. One ml Tris saturated phenol-chloroform was added and mixed gently, followed by centrifugation at 17,000 x g for 1 min. The aqueous layer was removed, washed twice with 1 ml chloroform, and transferred to a clean microcentrifuge tube. The sample was overlayed with 1 ml ice-cold 100 % ethanol and the DNA was spooled using a hooked Pasteur pipet. The spooled

DNA was washed by submerging in ice-cold ethanol, air dried for 5 minutes, and suspended overnight in 1 ml of Tris-EDTA buffer (10 mM Tris pH 8.0 and 1 mM EDTA). The concentration of recovered DNA was determined using a NanoDrop 2000 (Thermo Scientific,

Waltham, MA). Visualization by DNA gel electrophoresis ensured high-molecular weight

(>10,000 kb) DNA was abundant prior to sequencing. 140

Genomic DNA was submitted to the Georgia Genomics Facility (GGF) at the University of Georgia (Athens, GA) for paired-end (2 x 250 bp) sequencing using the Illumina MiSeq platform. DNA samples were fragmented and tagged with sequencing adapters using the Nextera

XT DNA sample preparation kit (Illumina, San Diego, CA). The sequencing data was processed and assembled by the Georgia Advanced Computing Resource Center (GACRC) at the

University of Georgia. Briefly, the raw sequencing data was cleaned up using Trimmomatic

(Usadel Lab, Max Planck Institute, Germany) with a read length cut-off of 100bp, resulting in

>300-fold coverage of the 4.95 Mb S. enterica LT2 genome (30). Trimmed reads were mapped to the published genome using Bowtie 2 (Source Forge). Variant calling was performed using the Genome Analysis Toolkit (Broad Institute, Cambridge, MA), and single nucleotide polymorphisms (SNPs) were identified using the Integrative Genomics Viewer (Broad Institute).

β-galactosidase assays. β-galactosidase assays were performed with modifications of previously described methods (31, 32). A 50-µL sample of overnight culture grown in NB medium with ampicillin (30 mg/L) was inoculated into 5 mL NB Amp30. The cultures were incubated at 37°C with shaking until OD600 = 0.5-0.7, at which time samples were removed and incubated with α-

MG (0.5%) or an equal volume ddH2O in a 96-well plate. After cells were incubated +/- α-MG for 45 min at 37°C with shaking, the OD600 of 175-µL cell samples was determined by a

SpectraMax 385 Plus microplate spectrophotometer (Molecular Devices, Sunnyvale, CA).

Twenty-µL samples of cells were then added to 80 µL permeabilization solution that contained

Na2HPO4 (100 mM), KCl (20 mM), MgSO4 (2 mM), hexadecyltrimethylammonium bromide

(1.6 mM), deoxycholic acid sodium salt (0.9 mM), and β-mercaptoethanol (77 mM). After at least 10 min incubation, 25 µL permabilized cell mixture was added to 150 µL substrate solution that contained Na2HPO4 (60 mM), NaH2PO4 (40 mM), o-nitrophenyl-β-D-galactoside (ONPG, 141

3.3 mM), and β-mercaptoethanol (39 mM). The reactions were incubated at 30°C and ONP product formation was monitored at A420 over time. Rates (ΔA420/min) were determined by fitting the data to a linear equation with outlier elimination in GraphPad Prism 6.0d (La Jolla, CA).

Miller units were calculated using the formula 1000*rate (ΔA420/min)/OD600/mL cell sample.

6.4 RESULTS AND DISCUSSION

Thi5p functions conditionally to complement a S. enterica thiC mutant strain. The THI5 gene was cloned into the pSU18 vector (pDM1336 and pDM1381) with expression from the lac promoter, which is constitutive in S. enterica. THI5 was unable to complement a S. enterica thiC mutant strain on minimal glucose medium (Table 6.3). Addition of 2 or 5 mM cAMP did not alter growth (data not shown), suggesting that catabolite repression was not preventing THI5 expression. Further analysis defined a suite of carbon sources that supported Thi5p-dependent growth, indicating the gene product was synthesized. The carbon sources that allowed Thi5p to satisfy the HMP-P requirement were ribose, mannose, xylose, fumarate and succinate (Table 6.3, depicted in Fig. 6.2). Glycerol and fructose allowed intermediate growth. Similarly to growth on glucose, Thi5p could not support thiamine-independent growth on other carbon sources, including galactose, gluconate, glucose-6-phosphate, fructose-6-phosphate, pyruvate, acetate and malate. Growth was compared in the same medium with and without thiamine, eliminating any effects each carbon source had on growth that were unrelated to thiamine biosynthesis. THI5 expression induced from pDM1381 (pBAD18S-THI5) with arabinose (0.02% or 0.2%) also did not allow growth on minimal glucose medium without thiamine, and followed a similar pattern of conditional complementation (data not shown), suggesting that the lack of function during growth on glucose was not due to low expression levels. 142

TABLE 6.3 Thi5p-dependent growth using different carbon sources.a

Carbon Source - Thiamine + Thiamine

Acetate (50 mM) 0.103 ± 0.003 0.595 ± 0.056

Fructose 0.231 ± 0.067 0.738 ± 0.003

Fructose-6-P 0.145 ± 0.008 0.649 ± 0.008

Fumarate (50 mM) 0.155 ± 0.010 0.211 ± 0.005

Galactose 0.126 ± 0.005 0.745 ± 0.002

Gluconate 0.112 ± 0.008 0.516 ± 0.005

Glucose 0.123 ± 0.018 0.755 ± 0.015

Glucose-6-P 0.121 ± 0.015 0.597 ± 0.005

Glycerol 0.253 ± 0.086 0.806 ± 0.006

Malate (40 mM) 0.122 ± 0.026 0.637 ± 0.011

Mannose 0.585 ± 0.087 0.743 ± 0.003

Pyruvate (50 mM) 0.117 ± 0.008 0.857 ± 0.010

Ribose 0.514 ± 0.033 0.667 ± 0.005

Succinate (20 mM) 0.191 ± 0.050 0.260 ± 0.075

Xylose 0.640 ± 0.009 0.732 ± 0.004

a Maximum OD650 during growth with shaking at 37°C of strain DM14419 (ΔthiC1225 ΔaraCBAD pTHI5). Data represents averages and standard deviations of three independent cultures.

143

Glucose PTS or glk Gluconate zwf Galactose G6P Ru5P pgi Xylose

Mannose F6P X5P R5P pfkAB Fructose FBP Ribose

E4P Glycerol DHAP G3P

PEP pykAF, ptsI

Pyr

Acetate Ac-CoA

OAA Citrate

Malate AKG

Fumarate Succinate

FIGURE 6.2 Carbon sources that support Thi5p-dependent growth in S. enterica. Carbon sources tested for supporting Thi5p-dependent growth in S. enterica are depicted in boxes based on whether each: supports growth (white fill); supports intermediate growth (grey fill); or does not support growth (black fill). The genes encoding the relevant enzymes are shown. Abbreviations: PTS, phosphoenolpyruvate (PEP):carbohydrate phosphotransferase system; G6P, glucose-6-phosphate; Ru5P, ribulose-5-phosphate; F6P, fructose-6-phosphate; X5P, xylulose-5- phosphate; FBP, fructose-1,6-bisphosphate; E4P, erythrose-4-phosphate; DHAP, dihydroxyacetone phosphate; G3P, glyceraldehyde-3-phosphate; PEP, phosphoenolpyruvate; Pyr, pyruvate; Ac-CoA, acetyl-coenzyme A; AKG, α-ketoglutarate; OAA, oxaloacetate.

144

Suppressor alleles of sgrR and ptsI allow Thi5p-dependent thiamine biosynthesis in minimal glucose medium. Mutations that allowed growth on minimal glucose medium were isolated to provide insight into the mechanisms required to allow sufficient Thi5p activity for growth. Derivatives of DM13623 (ΔthiC1225 ΔaraCBAD pTHI5) that allowed Thi5p-dependent growth on minimal glucose medium arose at a rate of ~1 X 10-7, but no spontaneous mutations were stably maintained. Four independent stable derivatives of DM13623 isolated from minimal glucose plates with DES were further analyzed. Lesions in ptsI (ptsI611) and sgrR (sgrR1 and sgrR2) were identified by a combination of genetic mapping, genome sequencing and targeted sequencing (Table 6.4). One mutation (zxx10175) has not yet been identified. These suppressor strains formed two classes based on growth pattern in minimal glucose medium with and without thiamine (Fig. 6.3). The first class, ptsI611 and zxx10175, had a pronounced growth defect on minimal glucose medium with thiamine. The second class, sgrR1 and sgrR2, had a slight defect in growth rate compared to the parent strain on minimal glucose medium with thiamine. The persistence of the growth defect in the presence of thiamine indicated the suppressor mutations caused metabolic defects independent of thiamine biosynthesis.

Reduced activity of phosphoenolpyruvate (PEP):carbohydrate phosphotransferase system

(PTS) Enzyme I facilitates Thi5p function. ptsI encodes Enzyme I (EI) of the phosphoenolpyruvate (PEP):carbohydrate phosphotransferase system (PTS), which concomitantly transports and phosphorylates many carbohydrates (including glucose, mannose and fructose) using PEP as the energy source (Fig. 6.4). Each PTS enzyme accepts the phosphoryl group on a histidine residue and then donates it to the next enzyme in the relay. EI and HPr (encoded by ptsH) are general PTS enzymes and are the first two proteins in the phosphorelay. For glucose transport, HPr then transfers the phosphoryl group to Enzyme IIA 145

TABLE 6.4 Alleles isolated that allow Thi5p-dependent growth in minimal glucose medium.

Nucleotide Allele Variant Encoded Substitution sgrR1 G1573A SgrRG525R sgrR2 C355T SgrRR119W ptsI611 G1082A Enzyme IR361H

146

A 1 B 1 650 650 OD OD

0.1 0.1

0 10 20 30 0 10 20 30 Time (h) Time (h)

FIGURE 6.3 Suppressor mutations allow Thi5p-dependent growth on minimal glucose medium and form two classes. The thiC pTHI5 strain and derivatives were grown in minimal glucose medium without (A) and with thiamine (B). The strains shown are derivatives of DM13623 and represent the following classes: parent (circles); sgrR1 (DM13715, inverted triangles); ptsI611 (DM13718, triangles). Growth was monitored by optical density at 650 nm with shaking at 37°C. Data are representative growth curves.

147

Inducer Exclusion

PEP EI HPr~P EIIAGlc (ptsI)(ptsH)(crr) Pyruvate EI~P HPr EIIAGlc~P Glc EIIBC Glucose (ptsG) - + Adenylate Glucose-6-P Cyclase -

+ sgrST SgrR SgrR SgrT

FIGURE 6.4. PEP:glucose phosphotransferase system (PTSGlc) and sugar-phosphate stress response in S. enterica. This schematic represents relevant aspects of the glucose-specific PTS and the sugar-phosphate stress response. PEP donates a phosphate group to EI. The phosphate group is transferred to histidine residues on HPr, then EIIAGlc, then EIIBCGlc, which concomitantly transports and phosphorylates glucose (reviewed in [Deutscher 2014]. When glucose-6-P accumulates, SgrR is activated through an unknown signal (star) [Vanderpool 2004; Vanderpool 2007]. Activated SgrR induces expression of sgrS [Vanderpool 2004]. sgrS mRNA destabilizes ptsG mRNA and encodes the small peptide SgrT, which inhibits EIIBCGlc post- translationally [Vanderpool 2004; Wadler 2007]. Dotted lines indicate the interactions are mediated through RNA. Relevant gene names are included in parentheses. Abbreviations: PEP, phosphoenolpyruvate; EI, Enzyme I; EIIAGlc, Enzyme IIAGlc; EIIBCGlc, Enzyme IIBCGlc.

148

(EIIAGlc, encoded by crr). EIIAGlc then transfers the phosphoryl group to Enzyme II BC

(EIIBCGlc, encoded by ptsG), which transports and phosphorylates glucose.

The strain containing ptsI611 was characterized to determine its effect on PTS EI activity. The ptsI611 allele isolated as allowing Thi5p-dependent growth encodes EI with a histidine substituted for Arg361 in the C-terminal domain important for dimerization (33). The

EI dimer is required for phosphorylation by PEP (34). The slow growth rate of a ptsI611 strain on minimal glucose medium was corrected by expression of wild-type ptsI (strain DM14242 had doubling time 5.72 ± 1.87 h when ptsI expression was not induced and 1.62 ± 0.01 with 0.1 mM

IPTG in minimal glucose medium containing thiamine). The phenotype of the THI5 strain containing ptsI611 was compared to the ptsI421::Tn10(Tc) strain. The strain containing ptsI421::Tn10(Tc) did not grow on mannose. This agreed with previous reports that ptsI mutants cannot use mannose as a carbon source (35) since mannose is a PTS sugar. The strain containing ptsI611 (DM14429) grew on mannose (doubling time 4.02 ± 0.58 h), although more slowly than the isogenic strain containing wild-type EI (DM14428; doubling time 2.55 ± 0.32 h). These results combined showed that ptsI611 encodes an EI variant with reduced activity.

Constitutive SgrR variants allow Thi5p function. sgrR (formerly yabN) encodes a transcription factor that coordinates the sugar-phosphate stress response in enteric bacteria (36,

37). When activated, SgrR increases expression of a number of genes including those encoding a sugar efflux pump (setA (38)), a glutamic-pyruvic transaminase (alaC, formerly yfdZ (37)) and a sugar phosphatase (yigL (39)). SgrR was defined for its activation of the small RNA sgrS (37), which destabilizes transcripts of genes encoding PTS transporters (ptsG (36) and manXYZ (40)).

The small RNA sgrS also encodes a small peptide, SgrT, that inhibits Enzyme IIBCGlc activity post-translationally (36, 41). 149

The sgrR1 and sgrR2 alleles isolated as allowing Thi5p-dependent growth were analyzed for their effects on the sugar-phosphate stress response. SgrR has a predicted N-terminal DNA- binding domain and C-terminal solute-binding domain (36); analysis of S. enterica SgrR in the

Interpro database (42) identifies the N-terminal binding domain as amino acids 5-118, with the solute binding domain beginning at amino acid 163. sgrR1 encodes a variant with a substitution in the predicted solute-binding domain of SgrR (SgrRG525R), while sgrR2 encodes a variant with a substitution immediately following the predicted DNA-binding domain (SgrRR119W).

The effect of the sgrR alleles on transcription from the sgrS promoter was tested to determine how they might affect glucose metabolism. The SgrR inducer has not been identified, but the non-metabolizable glucose analog α-methylglucoside (α-MG) is known to induce sugar- phosphate stress and SgrR activity (36). In the presence of α-MG, strains containing sgrR1 and sgrR2 had decreased transcription from sgrSp, which could be due to decreased specific activity or decreased protein stability (Fig. 6.5). In the absence of α-MG, strains containing sgrR1 and sgrR2 maintained transcription from sgrSp, suggesting they encode constitutive activators.

Deletion of sgrR or sgrS did not allow Thi5p-dependent growth on minimal glucose medium

(data not shown), suggesting the sugar-phosphate stress response is important for the growth enabled by sgrR1 and sgrR2. Thus, it is more likely that the constitutive activity of SgrRG525R and SgrRR119W is important for Thi5p-dependent thiamine biosynthesis on minimal glucose medium.

As described above, SgrRG525R and SgrRR119W contain substitutions in the solute-binding domain and the linker sequence, respectively. It is plausible that SgrR activation requires a conformational change in the interaction between the two domains. SgrRG525R could simulate binding of the inducer, while SgrRR119W could simulate the conformational change induced. 150

1000 s t i n u

r e l

l 500 i M

0 α-MG - + - + - + - + sgrR (WT) sgrR1 sgrR2 ΔsgrR

FIGURE 6.5 Effects of sgrR mutations on transcription from the sgrS promoter. β- galactosidase assays were performed on cultures in grown in NB medium containing ampicillin (30 mg/L) to mid-log phase and then incubated with and without α-MG (0.5%) for 45 minutes before permeabilization. The strain background contained deletions of thiC and araCBAD -1 -1 -1 (strains DM14531-4) The data are reported in Miller Units (1000 X A420 min OD600 ml ) and represent averages and standard deviations from three independent cultures. These results are representative of four independent experiments.

151

SgrRG525R and SgrRR119W could allow Thi5p-dependent thiamine biosynthesis on glucose by inhibiting production and activity of Enzyme IIBCGlc throughout growth on glucose as a carbon source, rather than only during the period of sugar-phosphate stress. Alternatively, SgrR activation of other genes (e.g. alaC, setA or yigL) could be important for allowing Thi5p activity on glucose.

Elimination of the PEP:glucose phosphotransferase system allowed Thi5p-dependent thiamine biosynthesis on glucose. Characterization of the isolated suppressor alleles suggested that Thi5p activity depended on reducing flux of PTS glucose transport and phosphorylation. We tested disruptions in genes encoding PTS enzymes to determine whether complete absence of the glucose PTS allowed Thi5p-dependent thiamine biosynthesis on glucose. Disruptions in ptsI

(EI), ptsH (Hpr) and ptsG (EIIBCGlc) also allowed Thi5p-dependent growth in minimal glucose medium (Fig. 6.5), consistent with the model that reduction in glucose transport and phosphorylation allowed Thi5p activity in the cell.

In contrast, a disruption in crr eliminating EIIAGlc did not allow Thi5p-dependent growth on glucose. In addition to its role in the glucose PTS phosphoryl group transfer, EIIAGlc is also a central regulator of carbon metabolism (reviewed in (43)): EIIAGlc~P activates adenylate cyclase while unphosphorylated EIIAGlc inhibits transport of non-PTS sugars (inducer exclusion).

Addition of cAMP (5 mM) did not rescue the thiC crr pTHI5 strain (data not shown), so it is unlikely that EIIAGlc~P activation of adenylate cyclase is the relevant activity. Further, disruption of ptsI or ptsH allowed Thi5p to function during growth on glucose but would prevent phosphorylation of EIIAGlc, suggesting the unphosphorylated EIIAGlc would be responsible for the required activity. Therefore EIIAGlc could be required for its role in inducer exclusion or another unidentified function. 152

A 1 B 1 650 650 OD OD

0.1 0.1

0 10 20 30 0 10 20 30 Times (h) Times (h)

FIGURE 6.6 Disruption of PTS allows Thi5p-dependent thiamine biosynthesis on minimal glucose medium. The thiC pTHI5 strain and derivatives were grown in minimal glucose medium without (A) and with (B) thiamine. The strains shown are derivatives of DM14419 and contain insertions in the following PTS genes: parent (circles); ptsI (triangles); ptsH (diamonds); crr (squares); ptsG (inverted triangles). Growth was monitored by optical density at 650 nm with shaking at 37°C. Data are representative growth curves.

153

Exogenous glucose or glycolytic intermediates inhibited Thi5p-dependent growth on ribose.

Because reduction of glucose transport/phosphorylation allowed Thi5p-dependent thiamine biosynthesis, we considered the hypothesis that glycolytic intermediates (such as glucose-6-P) were responsible for inhibiting Thi5p activity. This hypothesis was investigated by adding glycolytic intermediates to medium containing ribose as a carbon source. The THI5 strain

(DM14419) grew on ribose without thiamine (final OD650: 0.552 ± 0.003). However, growth was inhibited when the following metabolites were added at 4 mM (final OD650 without/with thiamine): glucose (0.106 ± 0.009/0.762 ± 0.009); glucose-6-P (0.106 ± 0.005/0.768 ± 0.031); or fructose-6-P (0.118 ± 0.008/0.778 ± 0.029). Inhibition by fructose-1,6-bisphosphate was not determined because addition of sufficient glucose-6-P (0.4 mM) for activation of the hexose-P transport system (44) inhibited growth. These results suggested that glycolytic intermediates were inhibitory to the Thi5p activity.

Disruption of glycolysis allowed Thi5p-dependent growth on glucose. Since glycolytic intermediates inhibited Thi5p-dependent growth, it was formally possible that glycolytic intermediates were directly inhibiting Thi5p activity. Therefore, glycolysis was genetically dissected to determine the relevant intermediates. This led to the finding that elimination of phosphoglucose isomerase (encoded by pgi) or phosphofructokinase A (encoded by pfkA, and responsible for ~95% of phosphofructokinase activity in E. coli (45)) allowed Thi5p-dependent growth on glucose (Table 6.5). Because pgi mutants accumulate glucose-6-P (G6P) and pfkA mutants accumulate fructose-6-P (F6P) (46), it is unlikely that G6P or F6P inhibits Thi5p activity directly. Previous reports found E. coli strains disrupted in pgi and pfkA had destabilized ptsG mRNA (46, 47), presumably through the sugar-phosphate stress response (36). Therefore pgi and pfkA disruptions could allow Thi5p-dependent growth by decreasing PTS-mediated glucose 154

TABLE 6.5 Thi5p-dependent growth on glucose in strains disrupted in glycolysis. a

Doubling Time Final Cell Yield Relevant - Thiamine + Thiamine - Thiamine + Thiamine genotype thiC pTHI5 ∞ 1.16 ± 0.01 0.112 ± 0.001 0.836 ± 0.013 pgi thiC pTHI5 7.71 ± 1.01 6.70 ± 0.53 0.656 ± 0.006 0.867 ± 0.012 pfkA thiC pTHI5 6.55 ± 1.32 4.94 ± 0.39 0.652 ± 0.074 0.895 ± 0.021 aDoubing time (h) and final cell yield (max OD650 after of strains DM14419, DM14443 and DM14540 during growth with shaking at 37°C. Data represents average and standard deviations of three independent cultures

155 transport/phosphorylation, similarly to the ptsI and sgrR suppressor alleles isolated. The fact that disruptions in pgi and pfkA allow the THI5 strain to grow on glucose suggests glycolytic flux, rather than accumulation of glycolytic intermeidates, may be inhibit Thi5p function.

Conclusions. S. cerevisiae HMP-P synthase Thi5p is capable of supporting thiamine- independent growth in the S. enterica thiC mutant strain under certain conditions. The finding that the S. enterica THI5 strain was unable to synthesize HMP-P using glucose as a carbon source was further investigated through suppressor analysis. A suppressor mutation in ptsI caused decreased PTS Enzyme I activity. Suppressor mutations in sgrR caused constitutive SgrR activation; to our knowledge this is the first report of constitutive SgrR variants. Comparison to

ΔsgrR suggested the constitutive activity was relevant for Thi5p-dependent thiamine biosynthesis, implicating SgrR-activated inhibition of PTS Enzyme IIBCGlc. Taken together, the data described support a model where reduction of PTS-mediated glucose transport allows

Thi5p-dependent growth on minimal glucose medium. The finding that growth on glucose inhibits Thi5p activity in S. enterica could have implications for in vitro studies on Thi5p. If the

Thi5p inhibition is direct and stable throughout purification, production of Thi5p in minimal glucose medium (as reported by Lai et al., 2012) could affect its activity in reconstituted assays.

The suppressor mutations isolated did not shed light on why Thi5p cannot support growth on glucose medium, but suppression by pgi and pfkA suggested direct inhibition by glucose-6-P or fructose-6-P was unlikely. This is consistent with the fact that Thi5p likely encounters glucose-6-P and fructose-6-P in the cytoplasm of fungi. Although there is some evidence suggesting thiamine biosynthesis occurs in the mitochondria (reviewed in (48)), the characterization of a mitochondrial TPP transporter (49) and report that the S. cerevisiae THZ-P 156 synthase is localized to the cytoplasm (50) suggest thiamine biosynthesis occurs in the cytoplasm. The diverse carbon sources that do not support Thi5p-dependent growth and requirement of Enzyme IIAGlc for PTS suppression of the glucose phenotype suggest Thi5p inhibition may be due to a complex regulatory mechanism. These findings emphasize the fact that metabolism is highly interconnected and metabolic pathways are not simply “plug-and-play” even if all required substrates are present.

6.5 ACKNOWLEDGEMENTS

We acknowledge Man Him (Sammy) Leung for construction of pDM1336 and isolation and mapping of the sgrR1 suppressor mutation and Bryan Jones for contributing to characterization of ptsI611. We thank Katherine Miller and Timothy Hoover for ΔptsH::Kn. The National

Institutes of Health grant GM47296 to DMD supported this work. LDP was supported by NSF through Graduate Research Fellowship grant DGE-0718123.

6.6 REFERENCES

1. Yura T. 1956. Evidence of nonidentical alleles in purine requiring mutants of Salmonella typhimurium. Publ. Carnegie Inst. 612:63-75.

2. Newell PC, Tucker RG. 1968. Biosynthesis of the pyrimidine moiety of thiamine. A new route of pyrimidine biosynthesis involving purine intermediates. Biochem. J. 106:279- 287.

3. Martinez-Gomez NC, Downs DM. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

4. Chatterjee A, Li Y, Zhang Y, Grove TL, Lee M, Krebs C, Booker SJ, Begley TP, Ealick SE. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

5. Tazuya K, Yamada K, Kumaoka H. 1993. Pyridoxine is a precursor of the pyrimidine moiety of thiamin in Saccharomyces cerevisiae. Biochem. Mol. Biol. Int. 30:893-899.

6. Tazuya K, Yamada K, Kumaoka H. 1989. Incorporation of histidine into the pyrimidine moiety of thiamin in Saccharomyces cerevisiae. Biochim. Biophys. Acta 990:73-79. 157

7. Ishida S, Tazuya-Murayama K, Kijima Y, Yamada K. 2008. The direct precursor of the pyrimidine moiety of thiamin is not urocanic acid but histidine in Saccharomyces cerevisiae. J. Nutr. Sci. Vitaminol. 54:7-10.

8. Wightman R, Meacock PA. 2003. The THI5 gene family of Saccharomyces cerevisiae: distribution of homologues among the hemiascomycetes and functional redundancy in the aerobic biosynthesis of thiamin from pyridoxine. Microbiology 149:1447-1460.

9. Tanaka K, Tazuya K, Yamada K, Kumaoka H. 2000. Biosynthesis of thiamin under anaerobic conditions in Saccharomyces cerevisiae. Biol Pharm Bull 23:108-111.

10. Lai R-Y, Huang S, Fenwick MK, Hazra A, Zhang Y, Rajashankar K, Philmus B, Kinsland C, Sanders JM, Ealick SE, Begley TP. 2012. Thiamin pyrimidine biosynthesis in Candida albicans: a remarkable reaction between histidine and pyridoxal phosphate. J. Am. Chem. Soc. 134:9157-9159.

11. Coquille S, Roux C, Fitzpatrick TB, Thore S. 2012. The last piece in the vitamin B1 biosynthesis puzzle: structural and functional insight into yeast 4-amino-5- hydroxymethyl-2-methylpyrimidine phosphate (HMP-P) synthase. J. Biol. Chem. 287:42333-42343.

12. Hall C, Brachat S, Dietrich FS. 2005. Contribution of horizontal gene transfer to the evolution of Saccharomyces cerevisiae. Euk. Cell 4:1102-1115.

13. Dougherty MJ, Downs DM. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

14. Newell PC, Tucker RG. 1968. Precursors of the pyrimidine moiety of thiamine. Biochem. J. 106:271-277.

15. Choi-Rhee E, Cronan JE. 2005. A nucleosidase required for in vivo function of the S- adenosyl-L-methionine radical enzyme, biotin synthase. Chem. Biol. 12:589-593.

16. Choi-Rhee E, Cronan JE. 2005. Biotin synthase is catalytic in vivo, but catalysis engenders destruction of the protein. Chem. Biol. 12:461-468.

17. Farrar CE, Siu KK, Howell PL, Jarrett JT. 2010. Biotin synthase exhibits burst kinetics and multiple turnovers in the absence of inhibition by products and product-related biomolecules. Biochemistry 49:9985-9996.

18. von Mering C, Zdobnov EM, Tsoka S, Ciccarelli FD, Pereira-Leal JB, Ouzounis CA, Bork P. 2003. Genome evolution reveals biochemical networks and functional modules. Proc. Natl. Acad. Sci. U. S. A. 100:15428-15433.

19. Jeong H, Tombor B, Albert R, Oltvai ZN, Barabasi A-L. 2000. The large-scale organization of metabolic networks. Nature 407:651-654. 158

20. Koenigsknecht MJ, Downs DM. 2010. Thiamine biosynthesis can be used to dissect metabolic integration. Trends Microbiol. 18:240-247.

21. Vogel HJ, Bonner DM. 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218:97-106.

22. Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260-296.

23. Schmieger H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-88.

24. Downs DM, Petersen L. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.

25. Way JC, Davis MA, Morisato D, Roberts DE, Kleckner N. 1984. New Tn10 derivatives for transposon mutagenesis and for construction of lacZ operon fusions by transposition. Gene 32:369-379.

26. Bartolome B, Jubete Y, Martinez E, de la Cruz F. 1991. Construction and properties of a family of pACYC184-derived cloning vectors compatible with pBR322 and its derivatives. Gene 102:75-78.

27. Guzman LM, Belin D, Carson MJ, Beckwith J. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.

28. Koop AH, Hartley ME, Bourgeois S. 1987. A low-copy-number vector utilizing β- galactosidase for the analysis of gene control elements. Gene 52:245-256.

29. Caetano-Anolles G. 1993. Amplifying DNA with arbitrary oligonucleotide primers. PCR methods and applications 3:85-94.

30. McClelland M, Sanderson KE, Spieth J, Clifton SW, Latreille P, Courtney L, Porwollik S, Ali J, Dante M, Du F, Hou S, Layman D, Leonard S, Nguyen C, Scott K, Holmes A, Grewal N, Mulvaney E, Ryan E, Sun H, Florea L, Miller W, Stoneking T, Nhan M, Waterston R, Wilson RK. 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852-856.

31. Griffith KL, Wolf RE. 2002. Measuring β-galactosidase activity in bacteria: cell growth, permeabilization, and enzyme assays in 96-well arrays. Biochim. Biophys. Acta 290:397- 402.

32. Zhang X, Bremer H. 1995. Control of the Escherichia coli rrnB P1 promoter strength by ppGpp. J. Biol. Chem. 270:11181-11189.

33. Brokx SJ, Talbot J, Georges F, Waygood EB. 2000. Enzyme I of the phosphoenolpyruvate:sugar phosphotransferase system. In vitro intragenic 159

complementation: the roles of Arg126 in phosphoryl transfer and the C-terminal domain in dimerization. Biochemistry 39:3624-3635.

34. Weigel N, Waygood EB, Kukuruzinska MA, Nakazawa A, Roseman S. 1982. Sugar transport by the bacterial phosphotransferase system. Isolation and characterization of enzyme I from Salmonella typhimurium. J. Biol. Chem. 257:14461-14469.

35. Saier MH, Jr., Young WS, 3rd, Roseman S. 1971. Utilization and transport of hexoses by mutant strains of Salmonella typhimurium lacking enzyme I of the phosphoenolpyruvate- dependent phosphotransferase system. J. Biol. Chem. 246:5838-5840.

36. Vanderpool CK, Gottesman S. 2004. Involvement of a novel transcriptional activator and small RNA in post-transcriptional regulation of the glucose phosphoenolpyruvate phosphotransferase system. Mol. Microbiol. 54:1076-1089.

37. Vanderpool CK, Gottesman S. 2007. The novel transcription factor SgrR coordinates the response to glucose-phosphate stress. J. Bacteriol. 189:2238-2248.

38. Sun Y, Vanderpool CK. 2011. Regulation and function of Escherichia coli sugar efflux transporter A (SetA) during glucose-phosphate stress. J. Bacteriol. 193:143-153.

39. Papenfort K, Sun Y, Miyakoshi M, Vanderpool CK, Vogel J. 2013. Small RNA-mediated activation of sugar phosphatase mRNA regulates glucose homeostasis. Cell 153:426-437.

40. Rice JB, Vanderpool CK. 2011. The small RNA SgrS controls sugar-phosphate accumulation by regulating multiple PTS genes. Nucleic acids research 39:3806-3819.

41. Wadler CS, Vanderpool CK. 2007. A dual function for a bacterial small RNA: SgrS performs base pairing-dependent regulation and encodes a functional polypeptide. Proc. Natl. Acad. Sci. U. S. A. 105:20454-20459.

42. Hunter S, Jones P, Mitchell A, Apweiler R, Attwood TK, Bateman A, Bernard T, Binns D, Bork P, Burge S, de Castro E, Coggill P, Corbett M, Das U, Daugherty L, Duquenne L, Finn RD, Fraser M, Gough J, Haft D, Hulo N, Kahn D, Kelly E, Letunic I, Lonsdale D, Lopez R, Madera M, Maslen J, McAnulla C, McDowall J, McMenamin C, Mi H, Mutowo-Muellenet P, Mulder N, Natale D, Orengo C, Pesseat S, Punta M, Quinn AF, Rivoire C, Sangrador-Vegas A, Selengut JD, Sigrist CJ, Scheremetjew M, Tate J, Thimmajanarthanan M, Thomas PD, Wu CH, Yeats C, Yong SY. 2012. InterPro in 2011: new developments in the family and domain prediction database. Nucleic Acids Res. 40:306-312.

43. Deutscher J, Ake FM, Derkaoui M, Zebre AC, Cao TN, Bouraoui H, Kentache T, Mokhtari A, Milohanic E, Joyet P. 2014. The bacterial phosphoenolpyruvate:carbohydrate phosphotransferase system: regulation by protein phosphorylation and phosphorylation-dependent protein-protein interactions. Microbiol. Mol. Biol. Rev. 78:231-256. 160

44. Winkler HH. 1970. Compartmentation in the induction of the hexose-6-phosphate transport system of Escherichia coli. J. Bacteriol. 101:470-475.

45. Fraenkel DG, Kotlarz D, Buc H. 1973. Two fructose 6-phosphate kinase activities in Escherichia coli. J. Biol. Chem. 248:4865-4866.

46. Morita T, El-Kazzaz W, Tanaka Y, Inada T, Aiba H. 2003. Accumulation of glucose 6- phosphate or fructose 6-phosphate is responsible for destabilization of glucose transporter mRNA in Escherichia coli. J. Biol. Chem. 278:15608-15614.

47. Kimata K, Tanaka Y, Inada T, Aiba H. 2001. Expression of the glucose transporter gene, ptsG, is regulated at the mRNA degradation step in response to glycolytic flux in Escherichia coli. EMBO J. 20:3587-3595.

48. Hohmann S, Meacock PA. 1998. Thiamin metabolism and thiamin diphosphate- dependent enzymes in the yeast Saccharomyces cerevisiae: genetic regulation. Biochim. Biophys. Acta 1385:201-219.

49. Marobbio CM, Vozza A, Harding M, Bisaccia F, Palmieri F, Walker JE. 2002. Identification and reconstitution of the yeast mitochondrial transporter for thiamine pyrophosphate. EMBO J. 21:5653-5661.

50. Faou P, Tropschug M. 2004. Neurospora crassa CyPBP37: a cytosolic stress protein that is able to replace yeast Thi4p function in the synthesis of vitamin B1. J. Mol. Biol. 344:1147-1157.

! 161

CHAPTER 7

CONCLUSIONS

The goal of this thesis work was to better understand how thiamine biosynthetic enzymes work in the context of the metabolic network. Thiamine pyrophosphate (TPP) is a required cofactor made of two independently synthesized moieties, 4-amino-5-(hydroxymethyl)-2- methylpyrimidine phosphate (HMP-P), and 4-methyl-5-(2-hydroxyethyl)-thiazole phosphate

(THZ-P). When this thesis work was initiated, thiamine biosynthesis in Salmonella enterica was well characterized through genetic and biochemical work in S. enterica and Escherichia coli. All bacterial thiamine biosynthetic enzymes had been identified (1-4), and most of their activities had been reconstituted in vitro (5-9). However, gaps remained in the mechanistic understanding of some S. enterica thiamine biosynthetic enzymes, including the HMP-P synthase ThiC and the

THZ-P biosynthetic enzyme ThiI. The biochemical activity of ThiC had been successfully reconstituted (6, 7), but the in vitro reaction required large amounts of ThiC protein and produced sub-stoichiometic quantities of HMP(-P). Thus, it was an open question whether ThiC was a true enzyme capable of catalytic turnover or whether it was a single-turnover enzyme. In addition to its requirement for THZ-P biosynthesis, ThiI is also required for biosynthesis of the tRNA photosensor 4-thiouridine (2, 10). The ThiI mechanism for 4-thiouridine biosynthesis was well characterized (11-16). Although ThiI was required for formation of the ThiS- thiocarboxylate in THZ-P biosynthesis in vivo (17), its role remained unclear because it was not required for in vitro reconstitution of ThiS-thiocarboxylate formation or THZ-P synthesis (5, 18). ! 162

In contrast to bacterial thiamine biosynthesis, the key thiamine biosynthetic enzymes in

Saccharomyces cerevisiae remained uncharacterized when this thesis work was initiated in 2009.

Genetic studies implicated THI4 in eukaryotic THZ-P synthesis and THI5 family genes in aerobic HMP-P biosynthesis (19, 20). Since then, in vitro reconstitutions of both Thi4p and

Thi5p activity were reported (21, 22). Thi4p was shown to be a suicide enzyme, using an active site cysteine as the sulfur source to THZ-P biosynthesis (21). Inactivated Thi4p accumulated in

S. cerevisiae at the same rate as THZ-P production in vivo (21). Thi4p is also associated with protection from oxidative stress and mitochondrial DNA damage in multiple fungal species (23-

25), leading to the hypothesis that Thi4p inactivated for thiamine biosynthesis could have another role in the cell (26). Interestingly, in vitro reconstitution of Thi5p activity found that

Thi5p similarly used an active site histidine as a substrate for HMP-P biosynthesis and only produced half-stoichiometric amounts of HMP-P (22). Their data suggested Thi5p was a single- turnover enzyme (22), implying it was also a suicide enzyme like Thi4p. In support of their assignment as suicide enzymes, Thi4p and Thi5p are among the most dominant proteins in the S. cerevisiae proteome under thiamine-limiting conditions (27). However, questions remain about the physiological relevance of the Thi5p single-turnover activity. Unlike Thi4p, there are no reports of in vivo accumulation of inactivated Thi5p, nor of links between Thi5p and other areas of metabolism. We reasoned that probing Thi5p function in the context of S. enterica might identify physiological factors relevant to Thi5p activity to better understand its function in vivo.

7.1 CONCLUSIONS

7.1.1 Physiological analysis can be more sensitive than biochemical analysis. In Chapter 2, the activity of ThiC variants was compared in vivo and in vitro to identify factors affecting ThiC activity in an attempt to improve HMP-P production in the in vitro ThiC assays. In vivo analysis ! 163 of conditional thiC mutants divided them into three classes based on their requirements for function (28). The first class only functioned in genetic backgrounds predicted to have increased purine biosynthetic flux, and therefore seemed to require increased AIR levels. The second class required exogenous methionine; genetic analysis determined that they required increased SAM levels. The third class required exogenous pantothenate, a precursor to coenzyme A (CoA), and thus seemed to require increased CoA levels. In vitro analysis of representatives of these mutant classes found no correlation between in vivo and in vitro function, suggesting the in vivo growth phenotypes were more sensitive to changes in metabolite levels than the in vitro ThiC assay.

7.1.2 ThiC is a catalytic enzyme capable of multiple turnovers. The ThiC assay was optimized in an attempt to determine whether ThiC was capable of catalytic turnover. Achieving catalytic turnover would be conducive to assaying the properties of the enzyme and further mechanistic characterization. The key differences between the ThiC assays in Chapter 2 and

Chapter 3 included in vitro reconstitution of the ThiC [4Fe-4S] cluster and HPLC purification of the cosubstrate SAM to approximately 99% purity. Other key improvements to the ThiC assay were previously reported, including use of physiologically-relevant reducing agents and synthesis of highly pure AIR (28). When tested with these improved conditions, ThiC produced

3X stoichiometric excess of the product HMP-P (29). The activity of other radical SAM enzymes was reportedly improved when product inhibition was alleviated by inclusion of methylthioadenosine nucleosidase (MTAN), a nucleosidase capable of hydrolyzing the radical

SAM cleavage product 5’-deoxyadenosine (MTAN) (30-32). When MTAN was included in the

ThiC assay, ThiC produced 5X stoichiometric excess of HMP-P. This was the first report of

ThiC “turning over” (i.e., converting substrate into product) multiple times, demonstrating that

ThiC is a catalytic enzyme rather than a single-turnover enzyme. Improvement of ThiC activity ! 164 and catalytic turnover may aid future investigation into the ThiC mechanism, which has not been fully elucidated.

7.1.3 ThiC has unique properties among radical SAM enzymes. The intramolecular rearrangement of AIR to HMP-P is one of the most complex reactions carried out by a single enzyme. As previously reported, ThiC has unique properties among radical SAM enzymes. First,

ThiC binds the [4Fe-4S] cluster with a CX2CX4C motif, which is a unique variation on the canonical radical SAM motif CX3CX2C (6, 33, 34). Second, labeling studies found that ThiC uses a novel radical SAM mechanism including two sequential abstractions of hydrogen atoms from the 5’-deoxyadenosyl radical (35). Finally, inhibition studies described in Chapter 3 demonstrated ThiC inhibition by the SAM-metabolites adenosine and methylthioadenosine

(MTA) (29), which do not inhibit the two other radical SAM enzymes that have been tested (31,

36). These findings suggest ThiC enzymes may be unique among the radical SAM enzymes. It is also possible that ThiC enzymes are the founding members of another class of radical SAM enzymes.

7.1.4 Only the ThiI rhodanese domain is necessary for THZ-P biosynthesis, uncovering misannotation of most “ThiI” enzymes. As mentioned above, there was no mechanistic explanation for the requirement of ThiI for thiamine biosynthesis. In Chapter 4, genetic analysis determined that only the C-terminal rhodanese domain of ThiI was required for THZ-P biosynthesis (37). Based on what was known about ThiS-thiocarboxylate formation and analogy to ThiI 4-thiouridine biosynthesis (11-16), the requirement for only the ThiI rhodanese domain led to a clear mechanistic model of ThiI function in THZ-P biosynthesis. In this mechanistic model, ThiI serves as a sulfur carrier protein by accepting a sulfur from IscS and donating it to activated ThiS via a persulfide (R-S-S-H) formed on Cys456 in the ThiI rhodanese domain. ! 165

Bioinformatic analysis found that fewer than one third of genes annotated as thiI encoded the rhodanese domain required for THZ-P biosynthesis, suggesting most thiI genes were misannotated. Studies reported since the manuscript in Chapter 4 was published have confirmed the misannotation: ThiI proteins in the firmicute Bacillus subtilis and the archaeon

Methanococcus maripaludis were not required for THZ-P biosynthesis. This study emphasized the “danger of annotation by analogy,” as described in an accompanying commentary (38). M. maripaludis encodes the Thi4p-type THZ-P synthase and presumably uses the Thi4p cysteine residue as the sulfur source for THZ-P biosynthesis. It is unclear whether B. subtilis THZ-P biosynthesis relies on an alternate rhodanese-like sulfur carrier domain/enzyme or an alternative sulfur mobilization strategy, such as use of small-molecule persulfides (discussed in Chapter 5 and 7.1.6).

7.1.5 Alternative sulfur mobilization to THZ-P biosynthesis requires the cysteine desulfhydrase CdsH in S. enterica. Results presented in Chapter 4 included the finding that exogenous cysteine allowed an alternate sulfur mobilization pathway to bypass the requirement for ThiI (37). The results of further investigation were consistent with the relevant molecules for cysteine rescue being the products of cysteine oxidation, cystine (Cys-S-S-Cys) and S- sulfocysteine (Cys-S-SO3). The alternative sulfur mobilization pathway required the major cysteine desulfhydrase in S. enterica, CdsH. The working model for cystine/S-sulfocysteine rescue of a thiI mutant is that a portion of cystine/S-sulfocysteine is reduced to cysteine in the cell; CdsH then desulfhydrates cysteine, releasing sulfide that interacts with cellular disulfides

(e.g. cystine) to form a persulfide (R-S-S-H) that can replace the ThiI persulfide in THZ-P biosynthesis. Alternative sulfur mobilization via sulfide-dependent persulfide formation could be ! 166 representative of THZ-P biosynthesis in organisms that synthesize THZ-P but do not encode the

ThiI rhodanese domain.

7.1.6 Saccharomyces cerevisiae Thi5p can function conditionally in S. enterica. Chapter 6 describes studies to probe the function of Thi5p in a non-native metabolic context. Because the substrates of Thi5p, histidine (perhaps Thi5p-derived), and pyridoxal-5’-phosphate (PLP), are present in the S. enterica metabolic network, we were surprised to find that Thi5p only functioned conditionally in S. enterica. Thi5p activity was sufficient to support thiamine- independent growth on some carbon sources but not others, including glucose. In order to identify metabolic factors inhibiting Thi5p function in S. enterica, suppressor mutants of the S. enterica THI5 strain were isolated on minimal glucose medium. Genetic analysis suggested that

Thi5p could function on minimal glucose medium when there was reduced glucose transport/phosphorylation by the phosphoenolpyruvate:carbohydrate phosphotransferase system

(PTS). Further investigation found that glycolytic intermediates inhibited Thi5p function even when other carbon sources were provided in the medium, but strains known to accumulate glycolytic intermediates allowed Thi5p function when glycolytic flux was disrupted. The working model is that glycolytic flux is inhibitory to Thi5p function. Future work will focus on developing a mechanistic explanation for the conditional Thi5p function in S. enterica. The conditional activity of Thi5p in S. enterica emphasizes that metabolic modules are not always interchangeable, because metabolic integration can have unexpected effects on enzyme function.

7.1.7 Enzyme activity is context dependent. Work described in this thesis, combined with previous reports, emphasizes the context dependence of enzyme activity. In the context of the cell, thiamine biosynthesis is sensitive to subtle changes in the metabolic network (reviewed in

(39)). Some of these sensitivities are mediated by changes in metabolite levels that affect enzyme ! 167 activity. For example, the activity of the bacterial HMP-P synthase ThiC is affected by changes in purine metabolism (28, 40-43), [Fe-S] cluster metabolism (44), methionine and SAM metabolism (28, 41, 42), and CoA metabolism (28, 45, 46). The activity of S. cerevisiae Thi5p is inhibited by unidentified cellular factors during expression in S. enterica (Chapter 6).

Likewise, in vitro reconstitutions of enzyme activity are sensitive to differences between the enzyme’s environments in a test tube v. the cell. For example, ThiC activity in vitro was limited by product inhibition (29), while in the cell 5’-deoxyadenosine is hydrolyzed by MTAN.

In a second example, the dispensability of ThiI in the in vitro reconstitution of ThiS- thiocarboxylate formation and THZ-P synthesis (5, 18) suggests there are key differences in sulfur chemistry in vitro v. in vivo. Conclusions from this dissertation contribute to the idea that cellular metabolism is best understood through synergistic analysis of data from physiological and biochemical studies.

7.2 FUTURE DIRECTIONS

7.2.1 Understanding how ThiC relates to the radical SAM superfamily. Results discussed in

Chapter 3 and 7.1.3 suggest ThiC may be unique among radical SAM enzymes. Although the unique cysteine motif is characteristic of ThiC enzymes generally (34), the other unique properties (use of two hydrogen abstractions and inhibition by adenosine/methylthioadenosine) were only described for the Caulobacter crescentus and S. enterica ThiC enzymes, respectively.

Future work could compare the enzymatic properties of ThiC homologs from a variety of organisms to characterize the conservation of these unique properties within the ThiC enzyme family. Determining whether the unique properties of ThiC enzymes are conserved among ThiC homologs would be a first step in determining whether these properties are conserved together ! 168

generally, i.e. is the CX2CX4C motif predictive of a radical SAM enzyme using two hydrogen abstractions and being sensitive to inhibition by adenosine/methylthioadenosine?

The unique properties of ThiC raise questions about the evolutionary history of ThiC compared to other radical SAM enzymes. The radical SAM superfamily is likely ancient (47).

Radical SAM catalysis has been hypothesized to be an evolutionary precursor to the highly evolved coenzyme adenosylcobalamin, due to their shared reliance on the 5’-deoxyadenosyl radical (48). The biosynthesis of many coenzymes requires radical SAM enzymes, including both moieties of thiamine, biotin, and lipoic acid. Coenzymes generally are hypothesized to be relics of the RNA world, based on their close relationships to nucleotides (such as NAD, CoA,

ATP) (49). Coenzymes may have expanded the chemical potential of early ribozymes; indeed, a

2013 study identified a ribozyme that used thiamine to decarboxylate a pyruvate-like suicide substrate (50). Furthermore, the integration of purine biosynthesis and HMP-P biosynthesis circumstantially suggests ThiC enzymes may be related to particularly ancient radical SAM enzymes. Are the unique properties of ThiC evolved from the common radical SAM enzyme ancestor or the only extant examples of ancestral traits?

Bioinformatic analysis could determine whether there are any other extant examples the

CX2CX4C motif observed in ThiC enzymes. Any non-ThiC radical SAM enzymes using the

CX2CX4C motif could be characterized for their use of two hydrogen abstractions and inhibition by adenosine/methylthioadenosine. If these properties were correlated in a divergent radical

SAM enzyme, it could suggest they may be ancestral properties of radical SAM enzymes.

Further understanding of how ThiC enzymes relate to the radical SAM superfamily generally could shed light on the evolutionary history of subclasses of radical SAM enzymes. ! 169

7.2.2 Characterizing the molecular mechanism of conditional Thi5p function in S. enterica.

Future work will characterize the mechanism of Thi5p inhibition during growth on some carbon sources in S. enterica. As reported in Chapter 6, genetic disruptions of the PTS allowed the THI5 strain to grow on glucose. The exception is the disruption of crr, which encodes PTS Enzyme

IIAGlc (EIIAGlc). Interestingly, lack of EIIAGlc also prevents Thi5p function during growth on mannose and ribose (unpublished data). Future work will focus on identifying patterns of metabolic differences between permissive and non-permissive carbon sources and determining the essential role of EIIAGlc in Thi5p function in S. enterica. Depending on the molecular mechanism of glucose inhibition, future analysis could include purification of Thi5p from E. coli cells grown on glucose v. a permissive carbon source to determine whether growth conditions affect Thi5p activity in vitro.

7.2.3 Characterizing the link between thiamine biosynthesis and Coenzyme A. The link between CoA and ThiC was identified under conditions of reduced purine biosynthetic flux, and therefore decreased AIR levels (45, 46). Thus, compromising the ThiC substrate uncovered the subtle effect of CoA levels on ThiC activity. As described in Appendix A, yggX panE mutant strains with increased oxidative stress and reduced CoA levels have subtle defects in both HMP-

P and THZ-P biosynthesis. This phenotype is reminiscent of strains with increased oxidative stress and defects in glutathione biosynthesis, which had defects in [Fe-S] cluster metabolism

(51). Compromised [Fe-S] cluster metabolism is expected to affect synthesis of both thiamine moieties due to the oxygen-labile [Fe-S] clusters in the radical SAM enzymes ThiC and ThiH in

HMP-P and THZ-P biosynthesis, respectively. Therefore, the preliminary finding that CoA levels can affect both HMP-P and THZ-P biosynthesis suggests CoA levels may be important radical SAM enzymes generally, perhaps for synthesis or maintenance of their [Fe-S] clusters. ! 170

To directly test the hypothesis that CoA is important for [Fe-S] cluster metabolism, ThiC and

ThiH could be purified from a yggX panE strain and analyzed for [Fe-S] cluster loading.

7.3 REFERENCES

1. Petersen LA, Downs DM. 1997. Identification and characterization of an operon in Salmonella typhimurium involved in thiamine biosynthesis. J. Bacteriol. 179:4894-4900.

2. Webb E, Claas K, Downs DM. 1997. Characterization of thiI, a new gene involved in thiazole biosynthesis in Salmonella typhimurium. J. Bacteriol. 179:4399-4402.

3. Webb E, Downs D. 1997. Characterization of thiL, encoding thiamin-monophosphate kinase, in Salmonella typhimurium. J. Biol. Chem. 272:15702-15707.

4. Vander Horn PB, Backstrom AD, Stewart V, Begley TP. 1993. Structural genes for thiamine biosynthetic enzymes (thiCEFGH) in Escherichia coli K-12. J. Bacteriol. 175:982-992.

5. Kriek M, Martins F, Leonardi R, Fairhurst SA, Lowe DJ, Roach PL. 2007. Thiazole synthase from Escherichia coli: an investigation of the substrates and purified proteins required for activity in vitro. J. Biol. Chem. 282:17413-17423.

6. Martinez-Gomez NC, Downs DM. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

7. Chatterjee A, Li Y, Zhang Y, Grove TL, Lee M, Krebs C, Booker SJ, Begley TP, Ealick SE. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

8. Reddick JJ, Kinsland C, Nicewonger R, Christian T, Downs DM, Winkler ME, Begley TP. 1998. Overexpression, purification and characterization of two pyrimidine kinases involved in the biosynthesis of thiamin: 4-amino-5-hydroxymethyl-2-methylpyrimidine kinase and 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate kinase. Tetrahedron 54:15983-15991.

9. Backstrom AD, Mcmordie RAS, Begley TP. 1995. Biosynthesis of thiamin I: the function of the thiE gene product. J. Am. Chem. Soc. 117:2351-2352.

10. Mueller EG, Buck CJ, Palenchar PM, Barnhart LE, Paulson JL. 1998. Identification of a gene involved in the generation of 4-thiouridine in tRNA. Nucleic Acids Res. 26:2606- 2610.

11. Kambampati R, Lauhon CT. 2000. Evidence for the transfer of sulfane sulfur from IscS to ThiI during the in vitro biosynthesis of 4-thiouridine in Escherichia coli tRNA. J. Biol. Chem. 275:10727-10730. ! 171

12. Palenchar PM, Buck CJ, Cheng H, Larson TJ, Mueller EG. 2000. Evidence that ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis, may be a sulfurtransferase that proceeds through a persulfide intermediate. J. Biol. Chem. 275:8283-8286.

13. Mueller EG, Palenchar PM, Buck CJ. 2001. The role of the cysteine residues of ThiI in the generation of 4-thiouridine in tRNA. J. Biol. Chem. 276:33588-33595.

14. Waterman DG, Ortiz-Lombardia M, Fogg MJ, Koonin EV, Antson AA. 2006. Crystal structure of Bacillus anthracis ThiI, a tRNA-modifying enzyme containing the predicted RNA-binding THUMP domain. J. Mol. Biol. 356:97-110.

15. You D, Xu T, Yao F, Zhou X, Deng Z. 2008. Direct evidence that ThiI is an ATP pyrophosphatase for the adenylation of uridine in 4-thiouridine biosynthesis. Chem. Biol. Chem. 9:1879-1882.

16. Mueller EG, Palenchar PM. 1999. Using genomic information to investigate the function of ThiI, an enzyme shared between thiamin and 4-thiouridine biosynthesis. Prot. Sci. 8:2424-2427.

17. Taylor SV, Kelleher NL, Kinsland C, Chiu HJ, Costello CK, Backstrom AD, McLafferty FW, Begley TP. 1998. Thiamin biosynthesis in Escherichia coli - Identification of ThiS thiocarboxylate as the immediate sulfur donor in the thiazole formation. J. Biol. Chem. 273:16555-16560.

19. Praekelt UM, Byrne KL, Meacock PA. 1994. Regulation of THI4 (MOL1), a thiamine- biosynthetic gene of Saccharomyces cerevisiae. Yeast 10:481-490.

20. Wightman R, Meacock PA. 2003. The THI5 gene family of Saccharomyces cerevisiae: distribution of homologues among the hemiascomycetes and functional redundancy in the aerobic biosynthesis of thiamin from pyridoxine. Microbiology 149:1447-1460.

21. Chatterjee A, Abeydeera ND, Bale S, Pai PJ, Dorrestein PC, Russell DH, Ealick SE, Begley TP. 2011. Saccharomyces cerevisiae THI4p is a suicide thiamine thiazole synthase. Nature 478:542-546.

22. Lai R-Y, Huang S, Fenwick MK, Hazra A, Zhang Y, Rajashankar K, Philmus B, Kinsland C, Sanders JM, Ealick SE, Begley TP. 2012. Thiamin pyrimidine biosynthesis in Candida albicans: a remarkable reaction between histidine and pyridoxal phosphate. J. Am. Chem. Soc. 134:9157-9159. ! 172

23. Faou P, Tropschug M. 2004. Neurospora crassa CyPBP37: a cytosolic stress protein that is able to replace yeast Thi4p function in the synthesis of vitamin B1. J. Mol. Biol. 344:1147-1157.

24. Machado CR, Praekelt UM, de Oliveira RC, Barbosa AC, Byrne KL, Meacock PA, Menck CF. 1997. Dual role for the yeast THI4 gene in thiamine biosynthesis and DNA damage tolerance. J. Mol. Biol. 273:114-121.

25. Ruiz-Roldan C, Puerto-Galan L, Roa J, Castro A, Di Pietro A, Roncero MI, Hera C. 2008. The Fusarium oxysporum sti35 gene functions in thiamine biosynthesis and oxidative stress response. Fungal Genet. Biol. 45:6-16.

26. Roach P. 2011. Biochemistry: Suicide of a protein. Nature 478:463-464.

27. Muller EH, Richards EJ, Norbeck J, Byrne KL, Karlsson KA, Pretorius GH, Meacock PA, Blomberg A, Hohmann S. 1999. Thiamine repression and pyruvate decarboxylase autoregulation independently control the expression of the Saccharomyces cerevisiae PDC5 gene. FEBS Lett. 449:245-250.

28. Palmer LD, Dougherty MJ, Downs DM. 2012. Analysis of ThiC variants in the context of the metabolic network of Salmonella enterica. J. Bacteriol. 194:6088-6095.

29. Palmer LD, Downs DM. 2013. The thiamine biosynthetic enzyme ThiC catalyzes multiple turnovers and is inhibited by S-adenosylmethionine (AdoMet) metabolites. J. Biol. Chem. 288:30693-30699.

30. Challand MR, Ziegert T, Douglas P, Wood RJ, Kriek M, Shaw NM, Roach PL. 2009. Product inhibition in the radical S-adenosylmethionine family. FEBS Lett. 583:1358- 1362.

31. Farrar CE, Siu KK, Howell PL, Jarrett JT. 2010. Biotin synthase exhibits burst kinetics and multiple turnovers in the absence of inhibition by products and product-related biomolecules. Biochemistry 49:9985-9996.

32. Choi-Rhee E, Cronan JE. 2005. A nucleosidase required for in vivo function of the S- adenosyl-L-methionine radical enzyme, biotin synthase. Chem. Biol. 12:589-593.

33. Sofia HJ, Chen G, Hetzler BG, Reyes-Spindola JF, Miller NE. 2001. Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res. 29:1097-1106.

34. Shisler KA, Broderick JB. 2012. Emerging themes in radical SAM chemistry. Curr. Opin. Struct. Biol. 22:701-710.

35. Chatterjee A, Hazra AB, Abdelwahed S, Hilmey DG, Begley TP. 2010. A "radical dance" in thiamin biosynthesis: mechanistic analysis of the bacterial ! 173

hydroxymethylpyrimidine phosphate synthase. Angew. Chem. Int. Ed. Eng. 49:8653- 8656.

36. Chirpich TP, Zappia V, Costilow RN, Barker HA. 1970. Lysine 2,3-aminomutase. Purification and properties of a pyridoxal phosphate and S-adenosylmethionine-activated enzyme. J. Biol. Chem. 245:1778-1789.

37. Martinez-Gomez NC, Palmer LD, Vivas E, Roach PL, Downs DM. 2011. The rhodanese domain of ThiI is both necessary and sufficient for synthesis of the thiazole moiety of thiamine in Salmonella enterica. J. Bacteriol. 193:4582-4587.

38. Bender RA. 2011. The danger of annotation by analogy: most "thiI" genes play no role in thiamine biosynthesis. J. Bacteriol. 193:4574-4575.

39. Koenigsknecht MJ, Downs DM. 2010. Thiamine biosynthesis can be used to dissect metabolic integration. Trends Microbiol. 18:240-247.

40. Yura T. 1956. Evidence of nonidentical alleles in purine requiring mutants of Salmonella typhimurium. Publ. Carnegie Inst. 612:63-75.

41. Newell PC, Tucker RG. 1968. Precursors of the pyrimidine moiety of thiamine. Biochem. J. 106:271-277.

42. Newell PC, Tucker RG. 1968. Biosynthesis of the pyrimidine moiety of thiamine. A new route of pyrimidine biosynthesis involving purine intermediates. Biochem. J. 106:279- 287.

43. Bazurto JV, Downs DM. 2014. Amino-4-imidazolecarboxamide ribotide directly inhibits coenzyme A biosynthesis in Salmonella enterica. J. Bacteriol. 196:772-779.

44. Dougherty MJ, Downs DM. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

45. Frodyma M, Rubio A, Downs DM. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar typhimurium. J. Bacteriol. 182:236-240.

46. Allen S, Zilles JL, Downs DM. 2002. Metabolic flux in both the purine mononucleotide and histidine biosynthetic pathways can influence synthesis of the hydroxymethyl pyrimidine moiety of thiamine in Salmonella enterica. J. Bacteriol. 184:6130-6137.

47. Frey PA, Hegeman AD, Ruzicka FJ. 2008. The radical SAM superfamily. Critical reviews in biochemistry and molecular biology 43:63-88.

48. Frey PA. 1993. Lysine 2,3-aminomutase: is adenosylmethionine a poor man's adenosylcobalamin? FASEB J. 7:662-670. ! 174

49. White HB. 1976. Coenzymes as fossils of an earlier metabolic state. J. Mol. Evol. 7:101- 104.

50. Cernak P, Sen D. 2013. A thiamin-utilizing ribozyme decarboxylates a pyruvate-like substrate. Nat. Chem. 5:971-977.

51. Gralnick J, Webb E, Beck B, Downs D. 2000. Lesions in gshA (Encoding gamma-L- glutamyl-L-cysteine synthetase) prevent aerobic synthesis of thiamine in Salmonella enterica serovar typhimurium LT2. J. Bacteriol. 182:5180-5187.

! 175

APPENDIX A

COENZYME A MAY PROVIDE ROBUSTNESS TO [FE-S] CLUSTER METABOLISM IN

SALMONELLA ENTERICA

A.1 INTRODUCTION

Coenzyme A (CoA) is an acyl-group carrier required for many metabolic processes including the tricarboxylic acid cycle, fatty acid oxidation and biosynthesis, and amino acid biosynthesis. In Salmonella enterica, CoA biosynthesis begins with biosynthesis of pantothenate

(vitamin B5). Pantothenate is made from the independently synthesized pantoate and B-alanine moieties. Pantoate is biosynthesized from the shared pantothenate/branched chain amino acid biosynthetic intermediate ketoisovalerate by the enzymes PanB and PanE (IlvC), while B-alanine is made from aspartate by PanD (Fig. A.1.A) (1, 2).!Pantothenate, cysteine and ATP are then converted to CoA by the enzymes encoded by coaA, coaBC, coaD, and coaE (Fig. A.1.A) (3-6).!

Previous work identified a link between cellular CoA levels and thiamine biosynthesis.

Thiamine pyrophosphate is a required cofactor, and is made of two independently synthesized moieties, 4-methyl-5-(2-hydroxyethyl)-thiazole phosphate (THZ-P) and 4-amino-5-

(hydroxymethyl)-2-methylpyrimidine phosphate (HMP-P). Mutations reducing CoA levels in the cell caused a conditional HMP-P auxotrophy in S. enterica (2, 7). These mutations were in the panE gene (formerly apbA), encoding ketopantoate reductase. panE mutant strains have a 90% reduction in CoA compared to the wild-type strain (8). A panE strain maintains 10% CoA compared to wild-type due to ketopantoate reductase activity by the branched chain amino acid ! 176 biosynthetic enzyme acetohydroxyacid isomeroreductase (IlvC) (9). Importantly, pantoate synthesis by IlvC is sufficient to prevent CoA auxotrophy in a panE mutant strain. Lesions in panE only result in an HMP-P auxotrophy when flux through the shared purine/HMP-P biosynthetic pathway is compromised by eliminating or inhibiting the enzyme PurF (2, 7, 8). In the presence of purines, expression of purF is repressed and the PurF enzyme is allosterically inhibited (10-12). The CoA effect was isolated specifically to the conversion of AIR to HMP-P in vivo (13), which we now know is catalyzed the HMP-P synthase ThiC (14, 15). ThiC is a member of the radical S-adenosylmethionine (SAM) superfamily of enzymes that use a [4Fe-4S] cluster to initiate radical catalysis (14, 15). CoA and acetyl-CoA do not exhibit any demonstrable effect on ThiC activity in vitro (16).

Previously described links between ThiC and [Fe-S] cluster metabolism suggest the ThiC

[Fe-S] cluster is particularly sensitive to oxidative stress in the cell (17). In S. enterica, the free thiol glutathione is the main cellular antioxidant. Thiamine auxotrophy is observed in strains that do not make glutathione and are defective in producing the small Fe-binding protein YggX (18-

20). In S. enterica, the reduced glutathione (GSH) is the predominant low-molecular weight thiol and serves as a cellular antioxidant. Glutathione was reported as important for maintaining [Fe-

S] cluster homeostasis in S. enterica (21). The glutathione-dependent thiamine auxotrophy required supplementation of both THZ and HMP (18), which was attributed to oxygen-labile

[4Fe-4S] clusters in the radical SAM enzymes ThiH (required for THZ-P biosynthesis) (18, 22) and ThiC (14, 15, 17). Importantly, the thiamine auxotrophy could be corrected by either GSH or

γ-glutamylcysteine, suggesting the thiamine requirement was caused by the loss of a cellular free thiol generally, rather than glutathione specifically (18). This study was initiated to characterize the link between CoA and thiamine biosynthesis. We started with the hypothesis that, as another ! 177

A PanE PanB (IlvC) PanD KIV Pantoate β-alanine Aspartate PanC

O O OH HO N H OH Pantothenate

CoaABCDE

NH2 O O N N O- O- HS O O O N N N P P O N H H OH O O

O OH Coenzyme A O P O- O- B PurFDTGI PurKECBH AIR IMP

[Fe-S] ThiC metabolism

H2N CoA N

PO N HMP-P COOH N ThiH, S ThiIFSG, IscS OP TPP THZ-P

FIGURE A.1 Relevant metabolic pathways in S. enterica. (A) Coenzyme A biosynthesis, (B) thiamine biosynthesis including previously described connections to [Fe-S] metabolism and CoA. Abbreviations: KIV, ketoisovalerate; AIR, 5-aminoimidazole ribotide; IMP, inosine monophosphate; HMP-P, 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate; THZ-P, 4- methyl-5-β-hydroxyethylthiazole phosphate; TPP, thiamine pyrophosphate.

! 178 dominant low-molecular weight thiol, CoA could provide robustness to the cellular antioxidant pool.

A.2 MATERIALS AND METHODS

Strains, media and chemicals. All strains used in this study are derived from S. enterica serovar

Typhimurium LT2, and are represented in Table A.1. All strains were generated for this study or were part of the laboratory collection. Tn10d(Tc) refers to the transposition-defective mini-Tn10 described by Way et al. (23). Rich medium is Difco nutrient broth (8 g/L) with NaCl (5 g/L).

Superbroth is tryptone (32 g/L), yeast extract (20 g/L), NaCl (5 g/L) with NaOH (0.05 N).

Minimal medium is no-carbon essential (NCE) medium supplemented with MgSO4 (1 mM), trace minerals (0.1 X) (24), and glucose (11 mM) or succinate (16.5 mM). Difco BiTek agar was added (15 g/L) for solid medium. When present in the medium, supplements were provided at the following concentrations: thiamine, 100 nM; HMP, 100 nM; THZ, 100 nM; adenine, 0.4 mM; methionine, 0.3 mM; pantothenate, 50 µM. Antibiotics were added at the following concentrations in rich and minimal media, respectively: chloramphenicol (Cm), 20 g/L, 5 g/l; ampicillin (Ap), 150 g/L, 30 g/L; tetracycline (Tc), 20 g/L, 10 g/L; kanamycin (Km), 50 g/L,

12.5 g/L. All chemicals were purchased from Sigma-Aldrich, St Louis, MO.

Transduction methods. The high-frequency generalized transducing mutant of bacteriophage

P22 (HT105/1, int-201) (25) was used for all transductional crosses. Transduction and subsequent purification was performed as previously described (7).

Growth analysis. Cells from overnight cultures in NB medium were pelleted and resuspended in an equal volume of saline (0.85% NaCl), and a 5-μl sample was used to inoculate 195 μl of the appropriate minimal medium in a 96-well plate. A microplate reader (model EL808, Bio-Tek

! 179

TABLE A.1 Strains used in this study.

Strain Number Genotype DM5647 yggX::Gm DM5990 yggX::Gm gshA102::MudJ DM12874 yggX::Gm panE::Cm DM14238 panE::Cm gshA102::MudJ DM13650 zxx-8029::Tn10d(Tc) panE::Cm DM13651 zxx-8029::Tn10d(Tc) thiC1128 panE::Cm DM13652 zxx-8029::Tn10d(Tc) thiC1129 panE::Cm DM13653 zxx-8029::Tn10d(Tc) thiC1146 panE::Cm

! 180

Instruments) was used to incubate strains at 37°C with shaking (medium). Cell density was measured as absorbance at 650 nm, and growth was reported as specific growth rates [μ =

ln(X/X0)/T]. Alternatively, a 200-μl sample of cell suspension was used to inoculate 5 mL of the appropriate minimal medium, and growth was monitored by A650 while strains were incubated at

37°C with shaking (200 rpm).

Assays for oxidative stress and [Fe-S] cluster metabolism. Sensitivity to H2O2 and streptonigrin were performed as described previously (26). Sensitivity to paraquat was performed as described previously (27). Aconitase and succinate dehydrogenase assays were performed as described previously (28).

A.3 RESULTS AND DISCUSSION

In a panE mutant strain, “adenine sensitive” ThiC variants cause pantothenate or thiamine auxotrophy. In a previously reported study, the ThiC variants (ThiCE281K, ThiCV267M, and

ThiCG92D) were designated “adenine sensitive” based on their inability to support thiamine- independent growth in the presence of adenine (29). Here, these variants were assessed for their ability to support thiamine-independent growth in a panE background. In the panE strain, the presence of these ThiC variants caused auxotrophy that was corrected by either pantothenate or thiamine (Table A.2). The fact that the pantothenate/thiamine auxotrophy was observed in

“adenine sensitive” ThiC variants builds on previous findings that the effect of CoA levels on

ThiC activity is subtle and depends on ThiC activity being compromised in another way. The conditional thiamine auxotrophy of a panE mutant strain was uncovered under conditions where

ThiC activity was compromised by restricting flux to the ThiC substrate, AIR (2, 7). The possibility that the effect of CoA levels was related to AIR levels specifically was supported by the above finding that “adenine sensitive” ThiC variants were sensitive to reduced CoA levels. ! 181

TABLE A.2 Growth of panE strains containing “adenine-sensitive” ThiC variants.a

ThiC Specific Growth Rate Final Cell Yield Strain variant Min Pant Thi Min Pant Thi 0.406 ± 0.671 ± 0.491 ± 0.749 ± 0.836 ± 0.711 ± DM13650 ThiCWT 0.009 0.004 0.006 0.010 0.027 0.011 0.028 ± 0.672 ± 0.501 ± 0.152 ± 0.891 ± 0.747 ± DM13651 ThiCE281K 0.002 0.019 0.005 0.001 0.019 0.023 0.009 ± 0.699 ± 0.534 ± 0.142 ± 0.877 ± 0.774 ± DM13652 ThiCV267M 0.003 0.018 0.008 0.012 0.017 0.001 0.007 ± 0.593 ± 0.534 ± 0.105 ± 0.862 ± 0.771 ± DM13653 ThiCG92D 0.006 0.002 0.008 0.003 0.016 0.012 a Strains were grown in NCE medium supplemented with glucose (11 mM) and the indicated additions. Min, minimal medium; Pant, pantothenate; Thi, thiamine. Growth rate is reported as μ

= ln(X/X0)/T, and the final cell yield is A650 after 12 h of growth. Data shown are the averages and standard deviations of three independent cultures.

! 182

Strains lacking yggX and panE display auxotrophy for both THZ and HMP. In order to determine whether the effect of CoA levels on thiamine biosynthesis was specific to AIR levels in the cell, the effect of a panE disruption was assessed in a strain background where ThiC activity was compromised in another way. Strains lacking YggX exhibit increased oxidative stress and conditional thiamine auxotrophy (18, 19, 27, 28, 30) attributable to the oxygen-labile

[Fe-S] clusters in ThiC and ThiH.

The nutritional requirements caused by a lesion in panE were assessed in a strain background lacking yggX. When grown in a soft agar overlay on minimal glucose medium, the panE yggX strain displayed auxotrophy satisfied by thiamine, pantothenate, or HMP and THZ in combination (Fig. A.2). Under these conditions, neither HMP nor THZ alone satisfied the nutritional requirement of the panE yggX strain, suggesting that both ThiC and ThiH were compromised. Strains lacking only panE or yggX were prototrophic (data not shown). The observed defect in THZ-P biosynthesis in the panE yggX strain is the first report of a link between CoA levels and THZ-P biosynthesis. The thiamine auxotrophy of the panE yggX strain was only observed using soft agar overlay growth assays; panE yggX was prototrophic when grown in liquid medium (in 5-mL volume or 96-well plate format) or when printed to solid minimal medium (data not shown). The specific growth conditions required to uncover the auxotrophy of the panE yggX strain suggests the metabolic defect is subtle.

The observed auxotrophy of the panE yggX strain is not due to a defect in glutathione biosynthesis. The nutritional requirements of the panE yggX strain was similar to the requirements of the gshA yggX strain (18, 19, 27). Therefore it was possible that the auxotrophy of the panE yggX strain was due to a defect in glutathione biosynthesis. However, growth of the panE yggX strain in soft agar overlay was not rescued by reduced glutathione (GSH) when a 1-μl ! 183

FIGURE A.2 The panE yggX strain of S. enterica is auxotrophic for both moieties of thiamine. Soft agar containing strain DM12874 (panE yggX) was overlaid on minimal glucose medium. Compounds were spotted in l-µL aliquots on the solidified agar as indicated, at the following concentrations: thiamine (B1), 10 µM; HMP and THZ; 100 µM, pantothenate, 1 mM. Abbreviations: Pant., pantothenate; HMP, 4-amino-5-(hydroxymethyl)-2-methylpyrimidine; THZ, 4-methyl-5-(2-hydroxyethyl)-thiazole.

! 184 sample of GSH (10 mM, 50 mM or 100 mM) was spotted (data not shown). Moreover, a panE gshA strain displayed a similar THZ and HMP auxotrophy when grown in soft agar overlay (data not shown). The additive nature of the defects caused by disruptions in panE and gshA suggests that panE and gshA affect metabolism through independent mechanisms. Interestingly, the panE gshA strain auxotrophy was satisfied by addition of thiamine, THZ and HMP together, or pantothenate, but not GSH (data not shown). The lack of rescue by GSH could be due to (1) the protective effect of glutathione from oxidative stress is dependent on sufficient CoA levels in the cell or (2) strains defective in CoA biosynthesis require more GSH than was added to overcome oxidative stress.

Disruptions in yggX and panE may contribute to increased oxidative stress by H2O2 through different mechanisms. The similarities between the nutritional requirements of the panE yggX strain and the gshA yggX strain suggested CoA could contribute to protection from oxidative stress like glutathione (27). The panE yggX strain was tested for sensitivity to compounds inducing different types of oxidative stress. The panE yggX strain showed a slightly increased sensitivity to H2O2 as compared to the panE or yggX strain, although much less significant than the gshA yggX strain (Fig. A.3). The slightly increased H2O2 sensitivity of the double mutant suggested CoA levels and YggX could contribute to resistance to oxidative stress through different mechanisms. The panE yggX strain did not show any increase in sensitivity to the superoxide generator paraquat or the antibiotic streptonigrin, which indicates available iron in the cell (31) (data not shown). In contrast, gshA yggX was previously reported to have increased sensitivity to paraquat and streptonigrin (26, 27). The difference in sensitivity to oxidizing agents between the panE yggX and gshA yggX strains is consistent with the hypothesis that metabolic defects caused by panE and gshA are independent. ! 185

100

10 yggXgshA yggX panE

% Survival yggXpanE 1 WT

0.1 0 1 2 Time (hrs)

FIGURE A.3 The panE yggX strain has slightly increased sensitivity to H2O2 stress. Percent survival was determined by viable cell assay before (0 h) and after (2 h) exposure to H2O2 (8 mm) of strains DM10000 (WT, closed circles); DM5647 (yggX, open triangle); DM12653 (panE, closed inverted triangle); DM12874 (panE yggX, open diamond); and DM5990 (gshA yggX, closed square). Data represent average and standard deviation of two independent cultures.

! 186

The panE yggX strain does not display defects in other [Fe-S] cluster enzymes under the conditions tested. In the absence of yggX, with disruptions in genes encoding [Fe-S] cluster metabolism enzymes have been shown to have defects in growth on succinate (due to the [Fe-S] cluster in succinate dehydrogenase), succinate dehydrogenase activity (SDH) and aconitase activity (21, 26-28, 32). The panE yggX strain was similarly analyzed for defects in [Fe-S] cluster metabolism. The panE yggX strain did not display defects in growth using succinate as a carbon source in most trials (four of five, data not shown). Likewise, the panE yggX strain did not display defects in aconitase or SDH activity in most trials (two of three, data not shown).

Taken together, these data suggest that any defect in [Fe-S] cluster metabolism is too subtle for consistent measurement.

Conclusions. The thiamine auxotrophy of the panE yggX strain described here suggests CoA levels also impact ThiH activity under certain conditions. Data presented within suggest the CoA effect on thiamine biosynthetic enzymes is not dependent on glutathione levels. Further analysis is required to determine whether CoA levels affect ThiC and ThiH specifically, radical SAM enzymes generally, or [Fe-S] cluster enzymes generally. The lack of consistently measurable effect on other [Fe-S] cluster enzymes could reflect specificity of the CoA effect to radical SAM enzymes, or could reflect the relative sensitivity of the [Fe-S] clusters in ThiC and ThiH.

If CoA levels affect [Fe-S] cluster enzymes generally, one possible mechanism is simply that the low-molecular weight CoA thiol (CoA-SH) acts as a cellular antioxidant and contributes to [Fe-S] cluster maintenance. One study of metabolite concentrations in E. coli reported GSH at

17 mM; CoA-SH was the second most abundant low-molecular weight thiol at 1.4 mM (33). The hypothesis that CoA-SH could serve as the major cellular antioxidant was previously proposed for Staphylococcus aureus, to explain the lack of glutathione in S. aureus (34); however, later ! 187 studies showed that the major cellular antioxidant in S. aureus is bacillithiol, a glutathione analog in Gram positive bacteria (35). If CoA-SH contributes to [Fe-S] cluster maintenance, it could explain why S. enterica maintains CoA levels at more than ten fold the level required to prevent auxotrophy (8). CoA-SH contributing to [Fe-S] cluster maintenance would provide metabolic robustness to the cellular antioxidant pools. Future studies could probe the relevance of this potential metabolic robustness in strains that do not synthesize glutathione.

A.4 REFERENCES

1. Cronan JE, Jr., Littel KJ, Jackowski S. 1982. Genetic and biochemical analyses of pantothenate biosynthesis in Escherichia coli and Salmonella typhimurium. J. Bacteriol. 149:916-922.

2. Frodyma ME, Downs D. 1998. The panE gene, encoding ketopantoate reductase, maps at 10 minutes and is allelic to apbA in Salmonella typhimurium. J. Bacteriol. 180:4757- 4759.

3. Song WJ, Jackowski S. 1992. Cloning, sequencing, and expression of the pantothenate kinase (coaA) gene of Escherichia coli. J. Bacteriol. 174:6411-6417.

4. Strauss E, Kinsland C, Ge Y, McLafferty FW, Begley TP. 2001. Phosphopantothenoylcysteine synthetase from Escherichia coli. Identification and characterization of the last unidentified coenzyme A biosynthetic enzyme in bacteria. J. Biol. Chem. 276:13513-13516.

5. Freiberg C, Wieland B, Spaltmann F, Ehlert K, Brotz H, Labischinski H. 2001. Identification of novel essential Escherichia coli genes conserved among pathogenic bacteria. J. Mol. Microbiol. Biotechnol. 3:483-489.

6. Mishra P, Park PK, Drueckhammer DG. 2001. Identification of yacE (coaE) as the structural gene for dephosphocoenzyme A kinase in Escherichia coli K-12. J. Bacteriol. 183:2774-2778.

7. Downs DM, Petersen L. 1994. apbA, a new genetic locus involved in thiamine biosynthesis in Salmonella typhimurium. J. Bacteriol. 176:4858-4864.

8. Frodyma M, Rubio A, Downs DM. 2000. Reduced flux through the purine biosynthetic pathway results in an increased requirement for coenzyme A in thiamine synthesis in Salmonella enterica serovar typhimurium. J. Bacteriol. 182:236-240.

9. Primerano DA, Burns RO. 1983. Role of acetohydroxy acid isomeroreductase in biosynthesis of pantothenic acid in Salmonella typhimurium. J. Bacteriol. 153:259-269. ! 188

10. Rolfes RJ, Zalkin H. 1988. Regulation of Escherichia coli purF. Mutations that define the promoter, operator, and purine repressor gene. J. Biol. Chem. 263:19649-19652.

11. Rolfes RJ, Zalkin H. 1988. Escherichia coli gene purR encoding a repressor protein for purine nucleotide synthesis. Cloning, nucleotide sequence, and interaction with the purF operator. J. Biol. Chem. 263:19653-19661.

12. Zhou G, Smith JL, Zalkin H. 1994. Binding of purine nucleotides to two regulatory sites results in synergistic feedback inhibition of glutamine 5-phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 269:6784-6789.

13. Allen S, Zilles JL, Downs DM. 2002. Metabolic flux in both the purine mononucleotide and histidine biosynthetic pathways can influence synthesis of the hydroxymethyl pyrimidine moiety of thiamine in Salmonella enterica. J. Bacteriol. 184:6130-6137.

14. Martinez-Gomez NC, Downs DM. 2008. ThiC is an [Fe-S] cluster protein that requires AdoMet to generate the 4-amino-5-hydroxymethyl-2-methylpyrimidine moiety in thiamin synthesis. Biochemistry 47:9054-9056.

15. Chatterjee A, Li Y, Zhang Y, Grove TL, Lee M, Krebs C, Booker SJ, Begley TP, Ealick SE. 2008. Reconstitution of ThiC in thiamine pyrimidine biosynthesis expands the radical SAM superfamily. Nat. Chem. Biol. 4:758-765.

16. Palmer LD, Downs DM. 2013. The thiamine biosynthetic enzyme ThiC catalyzes multiple turnovers and is inhibited by S-adenosylmethionine (AdoMet) metabolites. J. Biol. Chem. 288:30693-30699.

17. Dougherty MJ, Downs DM. 2006. A connection between iron-sulfur cluster metabolism and the biosynthesis of 4-amino-5-hydroxymethyl-2-methylpyrimidine pyrophosphate in Salmonella enterica. Microbiology 152:2345-2353.

18. Gralnick J, Webb E, Beck B, Downs D. 2000. Lesions in gshA (Encoding gamma-L- glutamyl-L-cysteine synthetase) prevent aerobic synthesis of thiamine in Salmonella enterica serovar typhimurium LT2. J. Bacteriol. 182:5180-5187.

19. Gralnick J, Downs D. 2001. Protection from superoxide damage associated with an increased level of the YggX protein in Salmonella enterica. Proc. Natl. Acad. Sci. U. S. A. 98:8030-8035.

20. Cui Q, Thorgersen MP, Westler WM, Markley JL, Downs DM. 2006. Solution structure of YggX: a prokaryotic protein involved in Fe(II) trafficking. Proteins 62:578-586.

21. Gardner PR, Fridovich I. 1993. Effect of glutathione on aconitase in Escherichia coli. Arch. Biochem. Biophys. 301:98-102.

22. Kriek M, Martins F, Leonardi R, Fairhurst SA, Lowe DJ, Roach PL. 2007. Thiazole synthase from Escherichia coli: an investigation of the substrates and purified proteins required for activity in vitro. J. Biol. Chem. 282:17413-17423. ! 189

23. Way JC, Davis MA, Morisato D, Roberts DE, Kleckner N. 1984. New Tn10 derivatives for transposon mutagenesis and for construction of lacZ operon fusions by transposition. Gene 32:369-379.

24. Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS. 1979. Methanogens: reevaluation of a unique biological group. Microbiol. Rev. 43:260-296.

25. Schmieger H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-88.

26. Thorgersen MP, Downs DM. 2008. Analysis of yggX and gshA mutants provides insights into the labile iron pool in Salmonella enterica. J. Bacteriol. 190:7608-7613.

27. Skovran E, Lauhon CT, Downs DM. 2004. Lack of YggX results in chronic oxidative stress and uncovers subtle defects in Fe-S cluster metabolism in Salmonella enterica. J. Bacteriol. 186:7626-7634.

28. Skovran E, Downs DM. 2000. Metabolic defects caused by mutations in the isc gene cluster in Salmonella enterica serovar typhimurium: implications for thiamine synthesis. J. Bacteriol. 182:3896-3903.

29. Palmer LD, Dougherty MJ, Downs DM. 2012. Analysis of ThiC variants in the context of the metabolic network of Salmonella enterica. J. Bacteriol. 194:6088-6095.

30. Vivas E, Skovran E, Downs DM. 2006. Salmonella enterica strains lacking the frataxin homolog CyaY show defects in Fe-S cluster metabolism in vivo. J. Bacteriol. 188:1175- 1179.

31. Bolzan AD, Bianchi MS. 2001. Genotoxicity of streptonigrin: a review. Mutat. Res. 488:25-37.

32. Skovran E, Downs DM. 2003. Lack of the ApbC or ApbE protein results in a defect in Fe-S cluster metabolism in Salmonella enterica serovar Typhimurium. J. Bacteriol. 185:98-106.

33. Bennett BD, Kimball EH, Gao M, Osterhout R, Van Dien SJ, Rabinowitz JD. 2009. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat. Chem. Biol. 5:593-599.

34. delCardayre SB, Stock KP, Newton GL, Fahey RC, Davies JE. 1998. Coenzyme A disulfide reductase, the primary low molecular weight disulfide reductase from Staphylococcus aureus. Purification and characterization of the native enzyme. J. Biol. Chem. 273:5744-5751.

35. Newton GL, Rawat M, La Clair JJ, Jothivasan VK, Budiarto T, Hamilton CJ, Claiborne A, Helmann JD, Fahey RC. 2009. Bacillithiol is an antioxidant thiol produced in Bacilli. Nat. Chem. Biol. 5:625-627.