Pyridine transport and assimilation in salmonella enterica

The goal of this work was to reveal how various pyridines are transported and assimilated in Salmonella enterica. In the first strategy, mutants were isolated that allow the use of nicotinamide mononucleotide (NMN) as a sole pyridine source when all known import routes are blocked. The major responsible mutations (pnuP*) affected a transmembrane-spanning protein, whose closest homolog is YegT, a putative nucleoside transporter. A minor class (pnuP*) affected the ydeA gene encoding a homolog of drug:H+ antiporters. It is suggested that PnuD and PnuP are both adapted transporters that have been altered to transport nicotinamide ribonucleoside (NmR). As seen for the PnuC transporter, assimilation of NMN using PnuD* or PnuP* depended on AphA and Nadl(T) functions. It is suggested that NMN is converted to NmR prior to import through PnuC, PnuD*, or PnuP*, and NmR is phosphorylated by Nadl(T) after entry. In the second strategy, mutants were isolated that improve quinolinic acid (Qa) assimilation. Mutants able to grow on a low concentration of Qa affected eight different loci. The affected genes were grouped into three categories based on their mode of action. Group I included four classes of mutations that recruited an adapted transporter: the pnuF* (yabN), pnuG* (opp), pnuH* (kefA), and pnul* mutations. It is suggested that wild-type PnuF and Opp may contribute to pyridine transport under some conditions. Group II comprised mutations in genes encoding enzymes involved in glycolysis: pgi, pfkA, and ptsG. It is proposed that these mutations enhance Qa assimilation by increasing the level of cAMP, which induces Qa transporter(s). Group III (pnuM) appeared to amplify the gabP gene encoding the GABA permease. This is the first report, in which the GABA transporter has been shown to transport Qa in any organism.

ENTERICA 1 degree of Philosophy May 2003 PYRIDINE TRANSPORT AND ASSIMILATION IN SALMONELLA ENTERlCA by Hotcherl Jeong A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor Department of Biology The University of Utah Hotcherl Copyright © Hotcher) Jeong 2003 All Rights reserved T H E U N I V E R S I T Y U T A H G R A D U A T E S C H O OL S U P E R V I S O R Y C O M M I T T E E A P P R O V A L satisfactory. }4 ; -zozz^ THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Hotcherl Jeong This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. David Blair 6irwood Casje Erik Jorgensen T H E U N I V E R S I T Y OF U T A H G R A D U A T E S C H O OL APPROVAL Hotcherl form tables, The Graduate School. Date Approved for the Major Department David Wolstenholme Department Chair Approved for the Graduate Council David S. Chapnic THE UN IVERSITY UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: I have read the dissertation of Hotcher! Jeong in its final foml and have found that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tahles, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory committee and is ready for submission to air: Supervisory Committee Q~otj6M!/~ -- Chapm Dean of The Graduate chool ABSTRACT assimilated in Salmonella enterica. In the first strategy, mutants were isolated that allow the use of nicotinamide mononucleotide (NMN) as a sole pyridine source when all pnuP*) transmembrane-spanning protein, whose closest homolog is YegT, a putative nucleoside H+ of NMN Nadl(Nadl(different opp), kefA), pnul* wild-type PnuF and Opp may contribute to pyridine transport under some conditions. Group II comprised mutations in genes encoding enzymes involved in glycolysis: pgi, pfkA, and ptsG. It is proposed that these mutations enhance Qa assimilation by increasing The goal of this work was to reveal how various pyridines are transported and known import routes are blocked. The major responsible mutations (pnuD*) affected a transporter. A minor class (pnuP*) affected the ydeA gene encoding a homolog of drug:H+ antiporters. It is suggested that PnuD and PnuP are both adapted transporters that have been altered to transport nicotinamide ribonucleoside (NmR). As seen for the PnuC transporter, assimilation ofNMN using PnuD* or PnuP* depended on AphA and NadI(T) functions. It is suggested that NMN is converted to NmR prior to import through PnuC, PnuD*, or PnuP*, and NmR is phosphorylated by NadI(T) after entry. In the second strategy, mutants were isolated that improve quinolinic acid (Qa) assimilation. Mutants able to grow on a low concentration of Qa affected eight different loci. The affected genes were grouped into three categories based on their mode of action. Group I included four classes of mutations that recruited an adapted transporter: the pnuF* (yabN), pnuG* (mill), pnuH* (kef A), and pnuI* mutations. It is suggested that Qg!, pikA, organism. v the level of cAMP, which induces Qa transporter(s). Group III (pnuM) appeared to amplify the gabP gene encoding the GABA permease. This is the first report, in which the GABA transporter has been shown to transport Qa in any organism. v Dedicated to My Father, Mother, Mother-in-law, and Wife, Jeongwon iv ix FIGURES xi of NAD Assimilation of NAD and NMN 2 Transport and Assimilation of Nm, Na, and Quinolinic Acid 5 References 7 OPENED P N U D 10 Abstract Methods ASSIMILATE Abstract Introduction Methods Results CONTENTS ABSTRACT ......................................................................................................... IV LIST OF TABLES ...................................................................... : ................... '...... IX LIST OF FIGURES............................................................................................... Xl Chapter 1. INTRODUCTION ...................................... ................................................... 1 Function and Biosynthesis ofNAD............................................................ 1 Assimilation ofNAD and NMN ............................................................... 2 Transport and Assimilation ofNm, Na, and Quinolinic Acid .................... 5 References................................................................................................. 7 2. A NEW ROUTE OF NICOTINAMIDE RIBOSIDE UPTAKE IS OPENED BY CHANGING A SINGLE AMINO ACID IN THE PNuD PROTEIN ....... 1 n Abstract..................................................................................................... 10 Introduction .... ...... ........ ......... ..... ....... .......... ............... ............ ................... 11 Materials and Methods........................................ .......... ....................... ...... 12 Results ....................................................................................................... 23 Discussion ...... ..................... ........ ...... .................. ......... .......... ................... 52 References ................................................................................................. 56 3. MUTATIONS THAT IMPROVE ABILITY TO ASSIMILATE QUINOLINIC ACID BY RECRUITING ADAPTED TRANSPORTERS ..... 62 Abstract..................................................................................................... 62 Introduction............................................................................................... 63 Materials and Methods............................. ....... .............................. ............. 65 Results....................................................................................................... 75 Discussion ................................................................................................. 99 References ................................................................................................. 103 MUTATIONS References gabP Results NmR of Qa viii 4. MUT A TIONS THAT ENHANCE QUINOLINIC ACID ASSIMILATION BY INCREASING cAMP LEVELS AND INDUCING RESIDENT TRANSPORTERS ................................. '" .................................................... 109 Abstract ..................................................................................................... 109 Introduction ............................................................................................... 110 Materials and Methods ............................................................................... 111 Results ....................................................................................................... 117 Discussion ... '" .................. '" ...................................................................... 130 References................................................................................................. 131 5. AMPLIFICATION OF THE gabP GENE ENCODING THE GABA PERMEASE ALLOWS GROWTH ON LOW QUINOLINIC ACID ............. 134 Abstract ..................................................................................................... 134 Introduction ............................................................................................... 135 Materials and Methods ............................................................................... 135 Results....................................................................................... ................ 140 Discussion ................................................................................................. 148 References ................................................................................................. 150 6. CONCLUSIONS ........................................................................................... 152 NMN and NrnR Assimilation Pathways ..................................................... 152 Transport and Assimilation ofQa .............................................................. 153 Vlll Table Page p n u P mutations of nadl pnuP* phenotype pnuP pyridines of pnuF, pnul assimilation mutations far pnuF gene 87 3.5 Sensitivity of the pnuG null mutants to pyridine analogues 93 3.6 Sequence changes in pnuH* mutations 96 3.7 Effects of a leuO mutation on Qa phenotypes of the PnuH* protein 97 3.8 Pnul proteins and its homologues 101 4.1 Strains used in this study of pnuL pnuK, and pnuL 112 4.2 Mutation types and sequence changes caused by pnuJ, K, and L mutations.. 120 4.3 Effects of a cya or crp mutation on different strains to assimilate Qa 129 5.1 Strains and plasmids used in this study of pnuM 136 LIST OF TABLES 2.1 Strains used in this study of pnuD and pnuP ........ ............................ ............. 13 2.2 Phenotypes of pncA pnuC suppressor mutants on NMN .. ............. ................ 23 2.3 Molecular and genetic comparison in the pnuD+ and its mutations................ 36 2.4 Effect ofnadI mutations on the pnuD* phenotype......................................... 38 2.5 Phenotypes of pncC and pnuD strains on different pyridines...... ..... .............. 43 3.1 Strains used in this study ofpnuF, pnuG, pnuH, and pnuI ............................ 66 3.2 Classification of the mutations elevating Qa transport and assimilation........ 71 3.3 Sequence changes in pnuF* mutations......................................................... 82 3.4 Effects of a fur mutation on growth phenotype and expression of the pnuF gene..................................... ................. .............................................. 87 3.5 Sensitivity of the pnuG null mutants to pyridine analogues ........ ....... ........... 93 3.6 Sequence changes in pnuH* mutations ........................................................ 96 3.7 Effects of a leuO mutation on Qa phenotypes of the PnuH* protein ............. 97 3.8 PnuI proteins and its homologues ................................................................ 101 4.1 Strains used in this study ofpnuJ, pnuK, and pnuL ...................................... 112 4.2 Mutation types and sequence changes caused by pnuJ, K, and L mutations .. 120 4.3 Effects ofa cya or £ill mutation on different strains to assimilate Qa ........... 129 5.1 Strains and plasmids used in this study ofpnuM .......................................... 136 5.2 Instability of 60-min markers in the pnuM mutants 5.3 Effect on growth phenotype in nadB Effect of a rpoS mutation growth phenotype pnuM mutants x 5.2 Instability of 60-min markers in the pnuM mutants ...................................... 146 5.3 Effect of gabP expression on Qa-growth phenotype in a nadB strain ............ 147 5.4 Effect of a!l!QS mutation on growth phenotype of pnuM mutants ................ 148 x O F Pag p n u P Typhimurium.... of MudJ p n u P p n u P gene serovar Typhimurium 32 2.5 The newly expanded NHS family within the MFS 33 2.6 Dependence of the PnuC and PnuD* proteins on NadlOT*) function 40 E. a n d Typhimurium rmuF 84 serovar Typhimurium 88 LIST OF FIGURES Figure 1.1 NAD metabolic pathway of Salmonella serovar Typhimurium ..................... 3 2.1 Genetic linkages of the pnuD region in Salmonella serovar Typhimurium .... 25 2.2 Relative location ofMudJ insertions in the pnuD locus ................................. 28 2.3 Nucleotide sequence and deduced amino acid sequence of the Salmonella serovar Typhimurium pnuD gene......... ............................................. ............ 29 2.4 Hydrophobicity plots for the PnuD and PnuC proteins of Salmonella serovar Typhimurium ............. ..................... ........ ..... .................................... 32 2.5 The newly expanded NHS family within the MFS ........................................ 33 2.6 Dependence of the PnuC and PnuD* proteins on NadI(T) function ............. 40 ' 2.7 Comparisons of the pnuD-glyA region in Salmonella serovar Typhimurium and E. coli ............. ................ ........ ......................................... 45 2.8 Physical map of the pnuP region in Salmonella serovar Typhimurium .......... 48 2.9 Sequence and a helical-wheel model for TMS5 of the PnuP protein .............. 49 2.10 Structures of substrate candidates for PnuD+ and PnuD* ............................... 55 3.1 Genetic and physical maps of the pnuF region in Salmonella serovar Typhimurium.... ...... ........................................................ ....... .......... ......... .... 77 3.2 Organization of the pnuF gene and its upstream region ................................. 79 3.3 The first and secondary structures of the pnuF upstream region .................... 84 3.4 Genetic and physical maps of the pnuG region in Salmonella serovar Typhimurium.......................... ...................... ............... .................. ..... ..... ..... 88 of pnuF Typhimurium pnul Typhimurium Pnul L Blnl Xbal, GABA 157 xii 3.5 Growth ofpnuF pnuG double mutants on 0.1 mM Qa ................................... 91 3.6 Cooperation of the PnuF and PnuG proteins ................................................. 92 3.7 Genetic and physical maps of the pnuH region in Salmonella serovar Typhimurium........ ....... ...... ........ ..... ................. ......... ....... ..... ........ ................ 95 3.8 Genetic and physical maps of the pnuI region in Salmonella serovar Typhimurium.................................. ... ................. ........... ............................... 98 3.9 A helical-wheel model for TMS11 of the PnuI protein .................................. 100 4.1 Genetic and physical maps of the pnuJ region in Salmonella serovar Typhimurium ................................................................................................ 118 4.2 Genetic and physical maps of the pnuK-pnuE region in Salmonella serovar Typhimurium ................................................................................................ 121 4.3 Genetic and physical maps of the pnuL region in Salmonella serovar Typhimurium ................................................................................................ 124 4.4 A model for how the pnuJ, K, and L mutations enhance Qa assimilation by increasing cAMP levels ........................................................................... 126 5.1 Digestion of genomic DNA of the pnuM strains by the enzymes BInI and Xba1, and separation of the fragments by PFGE ............................................ 142 5.2 The pnuM region in Salmonella serovar Typhimurium ................................. 143 5.3 Structures of GAB A and pyridines ............................................................... 149 6.1 A model for NMN assimilation pathways ..................................................... 154 6.2 Summary of Salmonella pyridine transport systems ...................................... 157 XlI CHAPTER 1 nucleotides, NADPH, of N A D + nucleotides, H2O2) OH*): + F e 3 + » + F e 2 + F e 2 + + H 2 0 2 » Fe3 + + OH' + nucleotides nucleotides N A D + CHAPTER! INTRODUCTION Function and Biosynthesis of NAD The pyridine nuc1eotides, NAD+/NADH and NADP+/NADPH, are universally used carriers of hydrogen and electrons, or reducing power, and act as coenzymes in hundreds of oxidation-reduction reactions in all organisms. In addition, the DNA ligase reaction of bacteria uses the high-energy phosphate bond ofNAD+ to activate the joining of two DNA ends (15). Although these are valuable functions of pyridine nuc1eotides, one reduced form (NADH) is dangerous because it contributes to oxygen toxicity by reducing iron and thereby allows the Fenton reaction, in which hydrogen peroxide (H20 2) reacts with iron to form the extremely reactive hydroxyl radical (OHO): NADH + Fe3+ - NAD+ + Fe2+ Fe2+ + H202 - Fe3+ + OHo + OH- (Fenton reaction) The valuable and dangerous roles of pyridine nuc1eotides suggest that mechanisms may exist to optimize the size of intracellular pools of pyridine nuc1eotides in response to varying environmental conditions. In Salmonella enterica serovar Typhimurium, NAD+ is produced by a de novo synthetic pathway and by a salvage pathway that recycles endogenous nicotinamide mononucleotide (NMN) and assimilates exogenous pyridines (Fig. 1.1) (14, 18). One mechanism for regulating de novo biosynthesis is feedback inhibition, of which the first enzyme (NadB) is the main target (2). Another mechanism is transcriptional repression of biosynthetic enzymes. The first two biosynthetic enzymes (nadB and nadA) and one recycling enzyme (pncB) are repressed during growth on a high concentration of exogenous nicotinamide (Nm) or nicotinic acid (Na) (5, 9). This control is mediated by the NadI repressor in response to the internal level of NAD. The nadl gene encodes a trifunctional protein, which acts as a transcriptional repressor (the R function) (26, 27), as an internal nicotinamide riboside (NmR) kinase (the T function) required for assimilation of exogenous NMN (J. Grose and J. R. Roth, unpublished data), and as a NMN adenylyl transferase (the A function) (13, 19), which has not been seen to be physiologically important in Salmonella (U. Bergthorsson, personal communication). Assimilation of NAD variety essential exogenous NAD cannot be used as a pyridine source, but is first degraded to NMN by NAD pyrophosphatase (PnuE) (3, 4, 17). The produced NMN can then be assimilated by 1.1). sequential pncA. pncB, 1 pnuC NMN concentration of 10 \iM. This route can be blocked by pncA mutations. 2 mononucleotide (NMN) and assimilates exogenous pyridines (Fig. 1.1 ) ( 14, 18). One. mechanism for regulating de novo biosynthesis is feedback inhibition, of which the first enzyme (Nad8) is the main target (2). Another mechanism is transcriptional repression of biosynthetic enzymes. The fi rst two biosynthetic enzymes (nadB and nadA) and one recycling enzyme (pocB) are repressed during growth on a high concentration of exogenous nicotinamide (Nm) or nicotinic acid (Na) (5, 9). This control is mediated by the NadI repressor in response to the internal level of NAD. The nad! gene encodes a trifunctional protein, which acts as a transcriptional repressor (the R function) (26, 27), as an internal ni cotinamide riboside (NmR) kinase (the T function) required for assimilation of exogenous NMN (J. Grose and J. R. Roth, unpublished data), and as a NMN adenylyl transferase (the A function) ( 13, 19), which has not been seen to be physiologically important in Salmonella CU. Bergthorsson, personal commWlication). ofNAD and NMN Salmonella serovar Typhimurium assimilates a vari ety of exogenous pyridine compounds and converts them to the essenti al cofactors, NAD and NADP. Intact rust two different routes (Fig. 1.1 ). One involves cleavage by a periplasmic glycohydrolase to yield Nm, which is then transported and converted to internal NAD by sequential activities of the pncA, pncS, nadD, and nadE gene products ( 1 , 4, 6, 8, 11). To provide a source of all pyridines (in a nadA-pnue mutant strain), this route requires an exogenous f.tM . FIG. NAD metabolic pathway of Salmonella serovar Typhimurium. The following N m 11. IA, iminoaspartate; l -pyrophosphate; 3 FIG. 1.1. NAD metabolic pathway of Salmonella serovar Typhimurium. The following enzymes are included: 1. L-Aspartate oxidase; 2. Qa synthetase; 3. PRPP-Qa phosphoribosyl transferase; 4. NaMN adenylyl transferase; 5. NAD synthetase; 6. NAD kinase; 7. DNA ligase; 8. NAD(P)ase; 9. NMN deamidase; 10. Nm deamidase; II. PRPP-Na phosphoribosyl transferase; 12. NAD glycohydrolase; 13. NAD pyrophosphatase; 14. Acid phosphatase (periplasmic); 15. NmR transporter; 16. NmR kinase. Abbreviations: DHAP, dihydroxyacetone phosphate; lA, irninoaspartate; Na, nicotinic acid; NaAD, nicotinic acid adenine dinucleotide; NaMN, nicotinic acid mononucleotide; Nm, nicotinamide; NMN, nicotinamide mononucleotide; NmR, nicotinamide ribonucleoside; PRPP, 5-phosphoribosyl-I-pyrophosphate; Qa, quinolinic acid. COOH T* NadB f* NadA ^TyP00" NadC H2N COOH HN COOHf Qa ( ^ A s P 1A DHAP PRPP NaMN ^ PncB|n Cytoplasm COOH NadD^v^cooH NadE ^ y ^ ^ NadK ^ NaAD NAD COOH Na PncA^io CONH2 Nm Nadl(T) NmR 1 6 • Adenosine-5'- t Nm NmR- 12 AphA NMN 14 PnuE|i3 COOH COOH ..., t ~ f NadA r.r-COOH NadC f")f"""" NadD f")f"""" ~f")f"""" ~ f")f"""'" 0N~ " 1 k\ r~r'N..JJ ~V ADS ~ AD6 ~ 0 0 l.-.~N Asp lA DHAP a PRPP N:i.1N 'N~~-' R;r ~-R "",-o-1;°-ro-c", ¥ COOH HN COOH Q I J<i2 1..fr}.J";).- t I..fr}.:r;,.- I ~ II II ~ Pneat11 OH OH ·2 O COOH 9 Lig 7 X NADP Na 8 PnCA+lO ~coo~ Adenosi ne-2',S'- ~CONfi2 ~ diphosphate V NadI(1-0' Nm .. 16 " NMN Inner membrane Periplasm Qa Na PnuC 15 NMN NmR "14 PnUEt '3 NAD NMN can also be assimilated by a second pathway, whose activity is seen in a nadB pncA double mutant. Recent results suggest that this route involves periplasmic removal of the phosphate of NMN by AphA prior to import by PnuC (U. Bergthorsson, Y. Xu, J. Sterneckert, B. Khodaverdian, H. Jeong, and J. R. Roth, unpublished data). The pnuC gene maps at 17.2 min promoter-distal to the nadA gene in an operon, whose transcriptional repression by NAD is mediated by the repressor domain (R) of NadI (26, 27). The second domain (T) of NadI appears to have a nicotinamide riboside (NmR) kinase activity required for NMN assimilation (25, 27) (J. Grose and J. R. Roth, unpublished data). Whereas nadl(R) mutants transport normally and express the operon constitutively, nadl(T) mutants are unable to assimilate NMN using the PnuC transporter but show normal repression control of the nadB, nadA-pnuC, and pncB genes (25, 27). Normal assimilation of exogenous NMN by PnuC (in a nadB pncA mutant) requires 0.1 mM NMN and involves a periplasmic phosphatase (AphA), the NmR transporter (PnuC), and an internal NmR kinase (Nadl(T)). Certain pnuC mutations allow import of intact NMN and thus make NMN assimilation independent of both AphA and Nadl(T) functions. T r a n s p o r t of Nm, a n d E. non-energy-diffusion process in E. coli (12, 16), Rowe et al. have shown that the Na uptake depends 5 NMN can also be assimilated by a second pathway, whose activity is seen in a nadB pncA double mutant. Recent results suggest that this route involves periplasmic removal of the phosphate of NMN by AphA prior to import by PnuC (u. Bergthorsson, Y. Xu, J. Sterneckert, B. Khodaverdian, H. Jeong, and J. R. Roth, unpublished data). The pnuC gene maps at 17.2 min promoter-distal to the nadA gene in an operon, whose transcriptional repression by NAD is mediated by the repressor domain (R) of NadI (26, 27). The second domain (T) of NadI appears to have a nicotinamide riboside (NmR) kinase activity required for NMN assimilation (25, 27) (J. Grose and J. R. Roth, unpublished data). Whereas nadI(R) mutants transport normally and express the operon constitutively, nadI(T) mutants are unable to assimilate NMN using the PnuC transporter but show normal repression control of the nadB, nadA-pnuC, and pncB genes (25, 27). Normal assimilation of exogenous NMN by PnuC (in a nadB pncA mutant) requires 0.1 mM NMN and involves a periplasmic phosphatase (AphA), the NmR transporter (PnuC), and an internal NmR kinase (NadI(T)). Certain pnuC mutations allow import of intact NMN and thus make NMN assimilation independent of both AphA and NadI(T) functions. Transport and Assimilation ofNm, Na, and Quinolinic Acid The simple pyridine bases, such as Nm, Na, and quinolinic acid (Qa), can be taken up by .E. coli and Salmonella serovar Typhimurium to provide an alternative to the de novo pathway of NAD. However, little is known about transport of these pyridines. While earlier studies suggest that Na might be taken up by a non-energy-requiring on the presence of 5-phosphoribosyl-l-PPi (PRPP)-Na phosphoribosyl transferase (PncB) and an energy source (20). The existence of the active transport of Na in E. coli was suggested by a Nm-binding protein isolated from osmotically shocked cells (7). The binding protein appears to be Nm deamidase (PncA) (6). Selection for resistance to the Nm and Na analogues (6-aminoNm and 6-aminoNa) yielded the pncA, pncB, and nadD mutations, but no transport-defective mutations (6, 10). This suggests that either there is no permease specific for Nm or Na, or else redundant transporters make it difficult to detect a mutational defect in any single uptake route. Quinolinic acid can also be utilized as a pyridine source. A very high concentration of Qa (10"2 M) is required to support growth of a nadB mutant on minimal glucose medium (24), whereas low levels (10"6 M) of simple pyridines (Nm or Na) can suffice. It remained unclear whether the requirement for a Qa concentration reflects poor uptake, or a poor K m of the PRPP-Qa phosphoribosyl transferase (NadC) activity. Results presented here suggest that uptake is limiting. In animals, Qa is known to activate selectively the subtype of neuronal glutamate receptors sensitive to N-methyl-D-aspartate (NMDA) (23). Since activation of NMDA receptors by Qa leads to neuronal damage, Qa can play a pathological role in neurodegenerative diseases, such as the AIDS (acquired immunodeficiency syndrome)- dementia and Huntington's disease (21, 22). These physiological and pathological roles of Qa have been intensively investigated. This dissertation presents studies on transport and assimilation of pyridine bases and nucleosides that have implications for many fundamental aspects of bacterial physiology. The experiments were initiated in hopes of identifying new genes involved in 6 on the presence of5-phosphoribosyl-l-PP, (PRPP)-Na phosphoribosyl transferase (PocB) and an energy source (20). The existence of the active transport of Na in E. coli was suggested by a Nm-binding protein isolated from osmotically shocked cells (7). The binding protein appears to be Nm deamidase (PncA) (6). Selection for resistance to the Nm and Na analogues (6-aminoNm and 6-arninoNa) yielded the pocA, poeB. and nadD mutations, but no transport-defective mutations (6, to). This suggests that either there is no permease specific for Nm or Na, or else redundant transporters make it difficult to detect a mutational defect in any single uptake route. Quinolinic acid can also be utilized as a pyridine source. A very high concentration of Qa (10.2 M) is required to support growth of a nadS mutant on minimal glucose medium (24), whereas low level s (10-6 M) of simple pyridines (Nm or Na) can suffice. It remained unclear whether the requirement for a Qa concentration reflects poor uptake, or a poor Km of the PRPP-Qa phosphoribosyJ transferase (NadC) activity. Results presented here suggest that uptake is limiting. In animals, Qa is known to activate selectively the subtype of neuronal glutamate receptors sensitive to N-methyl-D-aspartate (NMDA) (23). Since activation of NMDA receptors by Qa leads to neuronal damage, Qa can play a pathological role in neurodegenerative diseases, such as the AIDS (acquired immunodeficiency syndrome)- dementia and Huntington's disease (21, 22). These physiological and pathological roles of Qa have been intensively investigated. This dissertation presents studies on transport and assimilation of pyridine bases and nucleosides that have implications for many fundamental aspects of bacterial physiology. The experiments were initiated in hopes of identifying new genes involved in transport and assimilation of simple pyridines (Chapters 3, 4, and 5) and NMN (Chapter 2). n u c l e o t i d e mononucleotide-coli. Roth. J. Bacteriol. typhimurium. Curr. Microbiol. Curr. Microbiol. 10:237-242. 1981. nucleotide glycohydrolase, 1002-a n d W a r r e n . of NAD pncC Bacteriol 137:1165-7. Griffith, T. W., and F. R. Leach. 1973. The effect of osmotic shock on vitamin Escherichia coli. Biochem. Rechsteiner, I m p e r i a l , . Biol Chem J. lac of a Pt 7 transport and assimi lation of simple pyridines (Chapters 3, 4, and 5) and NMN (Chapter 2). References 1. Andreoli, A. J., T. W. Okita, R. Bloom, and T. A. Grover. 1972. The pyridine nucleotide cycle: presence of a nicotinamide mononucleotide-specific glycohydrolase in Escherichia coli . Biochem Biophys Res Commun 49:264-9. 2. Cookson, B. T., B. M. Oliveria, and J. R. Rotb. 1987. Genetic characterization and regulation of the nadB locus of Salmonella typhimurium. 1. Bacterial. 169:4285-4293. 3. Falconer, D. F., Spector, M. P., and J. W. Foster. 1984. Membrane association of NAD pyrophosphatase in Salmonella typhimurium . Curro Microbial. CurT. 4. Foster, J. W. ]981. Pyridine nucleot ide cycle of Salmonella typhimurium: in vitro demonstration of nicotinamide adenine dinucleotide glycohydrolase, nicotinamide mononucleotide glycohydroJase, and nicotinamide adenine dinucleotide pyrophosphatase activities. J Bacteriol 145: 1 002-9. 5. Foster, J. W., E. A. Holley-Guthrie, and F. Warren. 1987. Regulation ofNAD metabolism in Salmonella typhimurium: genetic analysis and cloning of the nadR repressor locus. Mol Gen Genet 208:279-87. 6. Foster, J. W., D. M. Kinney, and A. G. Moat. 1979. Pyridine nucleotide cycle of Salmonella typhimurium: isolation and characterization of pncA, pncB, and pnce mutants and utilization of exogenous nicotinamide adenine dinucleotide. J Bacteriol137: 11 65-75. transport in Arch. Biochern. Biophys. 159:658-663. 8. Hillyard, D., M. Rcchsteiner, P. Manlapaz-Ramos, J. S. Imperial, L. J . Cruz, and B. M. Olivera. 1981. The pyridine nucleotide cycle. Studies in Escherichia coli and the human cell line D98/AH2. J Bioi Chern 256:8491-7. 9. Holley, E. A., M. P. Spector, and J . W. Foster. 1985. Regulation of NAD biosynthesis in Salmonella typhimurium: expression of nad-Iac gene fusions and identification ofa nad regulatory locus. J Gen Microbiol 131 (PI 10):2759-70. Roth. 1983. resistant Bacteriol typhimurium. Bacteriol. 170:2113-metabolism III. in Biol. 5144-5149. NMN Microbiol Biotechnol 1:Germany. Natl. USA)* 57:1700-1971. Biol Salmonella-typhimurium 170:18. Penfound, T., and W. Foster. 1996. Biosynthesis and recycling of NAD, p. 19. Rappleye, A., and R. Roth. 1997. A TnlO derivative (T-POP) isolation with transport Escherichia coli. 21. Stone, T. W. 2001. Endogenous neurotoxins from tryptophan. Toxicon 39:61-73. 22. Stone, T. W. 2001. Kynurenines in the CNS: from endogenous obscurity to 8 10. Hughes, K. T., B. T. Cookson, D. Ladika, B. M. Olivera, and J. R. Roth. J 983. 6-Aminonicotinamide-resistant mutants of Salmonella typhimurium. J Bacterial 154:1126-36. 11. Hughes, K. T., B. M. Olivera, and J. R. Roth. 1988. Structural gene for NAD synthetase in Salmonella typhimuriwn. J. Bacterial. 170:211 3-2120. 12. McLaren, J., T. C. Ngo, and B. M. Olivera. 1973. Pyridine nucleotide metaboli sm in Escherichia coli. HI. Biosynthesis from alternative precursors in vivo. J. BioI. Chem. 248:5 144-5 149. 13. Mushegian, A. 1999. The Purloined Letter: bacterial orthologs of archaeal NMN adenylyltransferase are domains within multifunctional transcription regulator NadR. J Mol Microbial Biotechnoll: 127-8. 14. Olivera, B. M., K. T. Hughes, P. Cordray, and J. R. Roth. 1989. Aspects of NAD metabolism in prokaryotes and eukaryotes, p. 527, Jacobson, M. K. And E. L. Jacobson (Ed.). ADP ribose transfer reactions: Mechanisms and biological significance., vol. 16. Springer Verlag, New York, New York, USA and Berlin, Gennany. 15. Olivera, B. M., and I. R. Lehman. 1967. Diphosphopyridine nucleotide; a cofactor for the polynucleotide-joining enzyme from Escherichia coli. Proc. Nat! . Acad. Sci. (USA)' 57:1 700-1704. 16. Olivera, B. M., and R. Lundquist. 1971 . DNA synthesis in Escherichia coli in the presence of cyanide. J Mol Bioi 57:263-77. 17. Park, U. M., J. R. Roth, and B. M. Olivera. 1988. Salmonella-typhimurium mutants lacking NAD pyrophosphatase. J. Bacteriol. 170:3725-3730. PenrouDd, J. ofNAD, 721-30. In F. C. Neidhardt (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2 ed. American Society for Microbiology, Washington, D. C. C. aDd J. Roth_ for of insertions wi th conditional (tetracycline-dependent) phenotypes. J Bacteriol 179:5827-34. 20. Rowe, J. J., R. D. Lemmon, and G. J. Tritz. 1985. Nicotinic acid transpon in Microbios. 44: 169-184. 200 1. therapeutic importance. Prog Neurobiol 64: 185-218. CNS. Pharmacol Microbiol 20:typhimurium. Bacteriol. 1311-1320. of a Salmonella typhimurium. J. 'nadl typhimurium J. Bacteriol. 1310. 9 23. Stone, T. W., and M. N. Perkins. 1981. Quinolinic acid: a potent endogenous excitant at amino acid receptors in eNS. Eur J Pharmacal 72:411-2. 24. Tritz, G. J. 1974. Characterization of the nadR locus in Escherichia coli. Can J MicrobioI20:205-9. 25. Zhu, N., B. M. Olivera, and J. R. Roth. 1991. Activity of the NMN transport system is regulated in Salmonella ryphimurium. J. Bacterial. 173: 131 J -1 320. 26. Zhu, N., B. M. Olivera, and J. R. Roth. 1988. Identification ofa repressor gene involved in the regulation of NAD de novo biosynthesis in Salmonella Iyphimurium. J. Bacteriol. 170: 117-125. 27. Zhu, N., and J . R. ,Roth. 1991. The nadI region of Salmonella ryphimurium encodes a bifunctional regulatory protein. 1. Bacterial. 173: 1302-131 O. N EW PnuD minor pnuP*) closest homologues are YegT, a putative nucleoside transporter, and CscB, a sucrose permease. All of the three independent pnuP* mutations sequenced affected the same codon changing Arg52 to Ser, Cys, or His and hence suggest a specificity change. Null mutations in p n u P did not impair use of any pyridines tested, suggesting that the pnuD* mutations altered a normally unrelated transporter so as to allow import of NMN. Consistent with this idea, the p n u P gene was not regulated in response to pyridine levels or to mutations in the NadI repressor. A minor class of NMN users affected the ydeA gene, which encodes a homologue of drug:H+ antiporters; this mutation (pnuP*) occurred at motif C, the antiporter motif. It is suggested that PnuD and PnuP are both adapted CHAPTER 2 A NEW ROUTE OF NICOTINAMIDE RIBOSIDE UPTAKE IS OPENED BY CHANGING A SINGLE AMINO ACID IN THE PouD PROTEIN Abstract In a search for mutations that up-regulate mmor transporters of pyridines, mutants were isolated that allow use of nicotinamide mononucleotide (NMN) as a sole pyridine source when all previously known import routes are blocked. The major responsible mutations (pouD·) affected a transmembrane-spanning protein, whose homoJogues pennease. pouD· ArgS2 SeT, eys, pouD pouD· nonnally pnuD or to mutations the NadI repressor. A minor class of NMN users affected the ydeA W anti porters; at motif C, the antiporter motif. It is suggested that PnuD and PnuP are both adapted 11 transporters that have been mutationally altered to transport nicotinamide ribonucleoside (NmR). As seen for the normal pyridine-regulated PnuC transporter, assimilation of NMN using either PnuD* or PnuP* transporters depended on AphA and Nadl(T) functions. AphA is a periplasmic acid phosphatase that converts NMN to NmR, suggesting that NmR (rather than NMN) is the true substrate of PnuD* and PnuP*. The requirement for Nadl(T) kinase for assimilation by PnuD* or PnuP* suggests that this kinase is not directly involved in transport but rather plays an internal role in metabolism; the Nadl(T) function was previously thought to be in control of PnuC transporter activity. The predictions based on this work led to direct tests, which showed that Nadl(T) function is an internal NmR kinase (J. Grose and J. R. Roth, unpublished data). Thus NMN is converted to NmR prior to import through PnuC, PnuD*, or PnuP*, and assimilation by phosphorylation of the transported NmR. Salmonella enterica serovar Typhimurium assimilates a variety of exogenous pyridine compounds and converts them to the essential cofactors, NAD and NADP (46). The pathways for biosynthesis, recycling, and assimilation of NAD are diagrammed in Fig. 1.1. The only pyridine transporter identified thus far is PnuC, which is thought to recognize NmR (32, 66, 67, 69) (J. Grose and J. R. Roth, unpublished data). In hopes of identifying additional transporters of NMN or NmR, mutants were isolated that opened new routes of NMN assimilation. Mutants were isolated in a pncA pnuC strain that lacks both known routes of NMN assimilation. The new PnuD* and PnuP* mutants open a I I transporters that have been mutationally altered to transport nicotinamide ribonucleoside (NmR). As seen for the normal pyridine-regulated PouC transporter, assimilation of N MN using either PouO· or Poup· transporters depended on AphA and NadI(T) functions. AphA is a peripiasmic acid phosphatase that converts NMN to NmR, suggesting that NmR (rather than NMN) is the true substrate of PouO· and Poup·. The requirement for NadJ(T) kinase for assimilation by PnuD* or Poup· suggests that this kinase is not directly involved in transport but rather plays an internal role in metabolism; the NadI(T) function was previously thought 10 be in control of PnuC transporter activity. The predictions based on this work led to direct tests, which showed that Nadl(T) function is an intema1 NrnR kinase (1. Grose and J. R. Roth, unpublished data). Thus NMN is converted to NmR prior to import through PnuC, PnuD*, or PnuP*, and assimilation by phosphorylation of the transported NmR. Introduction ofNMN pncA.l2ill!C new transport route and restore ability to use NMN. The PnuD protein altered by these mutations appeared to be a transmembrane-spanning protein, whose closest homologues are YegT, a putative nucleoside transporter, and CscB, sucrose permease that are driven by proton-motive forces. The pnuP* mutations altered the ydeA gene. As for the standard PnuC-mediated route of NMN assimilation, use of the recruited mutant transporters (PnuD* and PnuP*) depended on the Nadl(T) function. Sequence analysis revealed that both PnuD and PnuP proteins belong to a transporter superfamily, the major facilitator superfamily (MFS) (36, 43, 52). MFS transporters act as uniporters, symporters, or antiporters of various solutes in response to chemiosmotic ion gradients. Most members of this family are 400 to 600 amino acid residues in length and possess either 12 or 14 putative transmembrane-spanners (TMSs). Thirty-eight families have been currently recognized within the MFS based on sequence similarity, where each of the families transports a distinct class of structurally related substrates. The amino acid sequences of MFS transporters include conserved amino acid sequence motifs, which are either ubiquitous within the MFS or family-specific. Methods Salmonella serovar Typhimurium strain LT2, whose genotypes are listed in Table 2.1. Two phage Mu derivatives were used to make lac operon fusions; these were Mud 1-8 Ampr ) Km") MudJ, TnlO TnlOdCm (Cmr ) (17) and TniOA16A17 (Tcr ) (62), referred to as TnlOdTc. 12 new transport route and restore ability to use NMN. The PnuD protein altered by these mutations appeared to be a transmembrane-spanning protein, whose closest homologues are YegT, a putative nucleoside transporter, and CscB, sucrose permease that are driven by proton-motive forces. The pnuP* mutations altered the ydeA gene. As for the standard PnuC-mediated route of NMN assimilation, use of the recruited mutant transporters (PnuD* and PnuP*) depended on the NadI(T) function. Sequence analysis revealed that both PnuD and PnuP proteins belong to a transporter superfamily, the major facilitator superfamily (MFS) (36, 43, 52). MFS transporters act as uniporters, symporters, or anti porters of various solutes in response to chemiosmotic ion gradients. Most members of this family are 400 to 600 amino acid residues in length and possess either 12 or 14 putative transmembrane-spanners (TMSs). Thirty-eight families have been currently recognized within the MFS based on sequence similarity, where each of the families transports a distinct class of structurally related substrates. The amino acid sequences of MFS transporters include conserved amino acid sequence motifs, which are either ubiquitous within the MFS or family-specific. Materials and Methods Bacterial strains and transposons. All bacterial strains used are derivatives of L T2, 1-(Ampf) (6, 26) and MudI1734 (Km) (7), and are referred to as MudA and MudJ, respectively. Two transposition-defective derivatives of transposon Tn10 were Cm) TnlOt-.16t-.17 Tc) TnlQdTc. 2.1. TR10000 TT14890 TT15564 TT15565 TT15573 TT15574 TT21997 TT418 TN858 TT315 TT15577 TT15585 LT2 pncA278: :TniOdCm nadA219: :A 1052f Kmr pnuC 3652::Tnl0dTc)] pncA278: :TnK)dCm nadA219: :Al 052fKmr pnuC aroG Tn!0dTc)] pnuD*135 Tn!0dCm nadA219::MudJ[A1052(TCmr aroG 36S2::Tnl0dTc)] pnuD*136 Tn!0dCm Al052fKmr pnuC zbhz3652::Tni0dTc)] pnuD*143 pncA278: :TniOdCm MudJ[A1052(Kmr pnuC Tnl0dTcY] pnuP*144 Tn!0dTc glvAlO glvA540::TnlO 801::Tnl0dTc purG1739::Tn!0 Tn!0dCm nadA219: :Al 052fKmr pnuC Tn!0dTc)] zff-3685::Tn!0dTc pnuD*136 Tnl0dCm nadA219::Al052fKmr aroG Tn!0dTc)] Tn!0dTc glvAlO Tn!0dCm nadA219::Al052(Kmr pnuC Tn!0dTc)] zff-3685::Tn!0dTc pnuD*136 glvAlO Tnl0dCm AlQ52rKmr pnuC aroG Tn!0dTc)] pnuD*136 glvA540::TnlO pncA278: :TniOdCm nadA219: :MudJ[Al 052rKmr pnuC Tn!0dTc)] A1879rzff-3685-pnuD) SGSCa 13 TABLE 2. 1. Strains used in this study of pnuD and pnuP Strain Genotype Source TRIOOOO Salmonella enterica serovar Typhimurium L T2 wild-type Lab collection TTI4890 pncA278::TnlQdCm nadA219::MudJ[llI052(Km'!lill!C Lab collection aroG zbh-3652::TnlQdTcl] TTI5564 pncA278::TnlQdCm nadA219::MudJ[llI052(Km'!lill!C This study areG zbh-3652::TnlQdTcl] pnuD'I35 TTI5565 pncA278::TnlQdCm nadA219: :MudJ[ II I 052(Km' pnuC This study areG zbh-3652::TnlQdTcl] pnuD'136 TT15572 pncA278::TnlQdCm nadA219::MudJ[llI052(Km'!lill!C This study aroG zbh-3652::TnlQdTcl] pnuD'I43 TTI5573 pncA278::TnlQdCm nadA219::MudJ[llWf(Km'!lill!C This study aroG zbh-3652::TnlQdTcl] pnuP'I44 TTl5574 zff-3685::TnlQdTc This study TT21997 glyAIO Lab collection TT418 glyA540::TnlQ Lab collection TN858 zff-801 ::TnlQdTc . SGSC' TT315 purG I 739::TnlQ Lab collection TTl 5577 pncA278::TnlQdCm nadA219::MudJ[lll 052(Km' !lill!C This study aroG zbh-3652::TnlQdTcl] zff-3685::TnIOdTc pnuD'TT20183 pncA278::TnlQdCm nadA2I 9::MudJ[lll 052(Km' pnuC This study areG zbh-3652::TnlQdTcl] zff-3685::TnlQdTc glyAIO TT20184 pncA278::TnlQdCm nadA219: :MudJ[lll 052(Km'!lill!C This study aroG zbh-3652::TnlQdTcl] zff-3685::TnlQdTc pnuD'glyAIO TTI5585 pncA278::TnlQdCm nadA219::MudJ[llI052(Km'!lill!C This study areG zbh-3652::TnlQdTcl] pnuD'I36 glyA540::TnlQ TT20186 pncA278::TnIOdCm nadA219::MudJ[llWf(Km'!lill!C This study aroG zbh-3652::TnlQdTcl]llJ.li2(zff-3685-pnuDl Source Tnl0dCm A1052fKmr Tn!0dTc)] A1890fpnuD-glvA) pncA278::Tnl0dCm A1052aCmr zbh=3652::Tnl0dTc)] pnuD*A1904(;glvA-purG) TT15589 Tnl0dCm A1081(aroG) DUP1086[(nMB499)*MudA*(guaI563)] TT15590 Tnl0dCm A1081(DUP1086[(pnuD*glvA540::TnlO nadB499)*MudA*fpnuD+ Tni0dCm Al 052f Kmr pnuC study aroG Tnl0dTc)] pnuD261::pncA278: :Tnl0dCm A 1052fKmr Tn!0dTc)] pnuD262::MudJ pncA278: :TnI0dCm nadA219: :A1052(Kmr study Tnl0dTc)] pnuD263::MudJ Tnl0dCm A1052rKmr pnuC This study 3652::Tn!0dTc)] MudJ pncA278::Tni0dCm Al 052rKmr pnuC This study Tn!0dTc)] csiEl::MudJ pncA278::Tnl0dCm nadA219::A1052rKmr pnuC This study Tnl0dTcY] pnuD*TT15620 TT15621 Tni0dCm Al 052fKmr pnuC Tnl0dTcVl pnuD*serB1463::Tn!0 Tni0dCm MudJ[A 1052rKmr aroG Tnl0dTc)] pnuD*136 serB1463::Tn!0 R+14 TABLE 2.1. continued Strain Genotype TT20189 pncA278::TnlQdCm nadA219::MudJ[ill 052(Km' pnuC This study aroG zbh-3652::TnlQdTc)] ilI890(pnuD-gh:A) TT20180 pncA278: :TnlQdCm nadA219::MudJ[iI I 052(Km' pnuC This study aroG zbh-3652::TnlQdTc)] pnuD'136 t.1904(gh:A-purG) TTl5589 pncA278::TnlQdCm ilI081(nadA- arnG) This study DUPI 086[(nadB499)'MudA '(gua-563)] TTl 5590 pncA278::TnlQdCm ilI081 (nadA- aroG) This study DUPI086[(pnuD'136 glyA540::TnlQ nadB499)'MudA '(pnuD+ gly+ gua-563)] TT20190 pncA278::TnlQdCm nadA219::MudJ[ill 052(Km' Jllil!C This study arnG zbh-3652::TnlQdTc)] pnuD261 ::MudJ TT20191 pncA278::TnlQdCm nadA219::MudJ[ill 052(Km' pnuC This study aroG zbh-3652::TnlQdTc)] pnuD262::MudJ TT20192 pncA278::TnlQdCm nadA219::MudJ[il I052(Km' pnuC This study aroG zbh-3652::TnlQdTc)] pnuD263::MudJ TT20193 pncA278::TnlQdCm nadA219::MudJ[il l 052(Km' Jllil!C aroG zbh-3652 ::TnlQdTc)] pnuD264::MudJ TT20194 pncA278: :TnlQdCm nadA219::MudJ[iI I 052(Km' Jllil!C study aroG zbh-3652::TnlQdTc)] csiEI::MudJ TT15586 pncA278: :TnlQdCm nadA2 19: :MudJ[iI 1 052(Km' Jllil!C aroG zbh-3652::TnlQdTc)] pnuD'136 pnuD145 TTI5620 pncA278::TnlQdCm nadA219::MudJ[ilJ..Q22(Km' Jllil!C This study aroG zbh-3652::TnlQdTc)] pnuD'136 serB 1463::TnlQ TTl5621 pncA278::TnlQdCm nadA219::MudJ[t.1052(Km' pnuC This study arnG zbh-3652::TnlQdTc)] pnuD'I36 serBI463::TnlQ nadI312(R+T") TT15622 Tnl0dCm A1052(Kmr pnuC Tnl0dTc)] pnuD*136 serB1463::Tn!nadI331(R-T+) TT15623 pncA278::nadA219: :A 1052f K m r aroG Tnl0dTc)] pnuD*136 serB1463::Tn!0 R"T-) TT15624 pncA278::Tnl0dCm nadA219::A1052(Kmr Tnl0dTc)] pnuD*serB1463::Tn!0 nadI5iI(RsT-) TT20241 AnadB103 pnuD261::TT14938 AnadB103 pnuC104::MudJ pnuD261::pnuC103::Mud-Cm pnuD261::pnuC103::Mud-Cm nadA219::A1052(Kmr pnuC 3652::Tnl0dTcV] pnuD*136 nadI563::R'T-) TT9923 AnadB103 TT15148 Ai081(nadA-argG) AI081(nadA-arpG) pnuD261::MudJ TT12990 AnadB103 pncA278::Tnl0dCm TT22245 AnadB103 pncA278::Tnl0dCm pnuD261::MudJ pncB165::Tn!0 A108irnadA-aroG) TT22247 pncB165::Tn!0 A1081(nadA-aroG) pnuD261::MudJ pncA278::Tnl0dCm nadA219::MudJ[A1052fKmr pnuC Tn!0dTc)] vdeA137::Tn!0dTc pnuP*144 study This study This study This study This study 15 TABLE 2.1. continued Strain Genotype Source TTl 5622 pncA278::TnlQdCm nadA219::MudJ[td052(Km' Il!!l!C This study aroG zbh-3652::TnlQdTc)] pnuD' I36 serB 1463 ::Tn I 0 nadI33 I (R"T) TTI5623 pncA278: :TnlQdCm nadA219::MudJ[I'.lQ2l(Km' pnuC This study !![QQ zbh-3652::TnlQdTc)] pnuD'I36 serB I 463::TnlQ nadI322(Ri) TTl 5624 pncA278: :TnlQdCm nadA2 19::MudJ[I'. 1 052(Km' pnuC This srudy aroG zbh-3652::TnlQdTc)] pnuD'136 serBI463::TnlQ nadl511 (R") TT2024I "'nadBl03 pnuD261 ::MudJ This study TT I4938 "'nadB l03 pnuCI04: :MudJ Lab collection TT20240 pnuD261 ::MudJ This study TT22862 pnuCI03::Mud-Cm This study TT22865 pnuD261 ::MudJ pnuCI03::Mud-Cm TT23093 pncA278::TnlQdCm nadA2I 9::MudJ[Lac+ 1'.1 052(Km' This study Il!!l!C aroG zbh-3652: :TnlQdTc)] pnuD' I36 nadJ563::MudJ(Ri) MadBI03 Lab collection TTI5148 "'1081 (aroG) Lab collection TT22244 "' I 08 1 (aroG) pnuD26I ::This study TTl 2990 "'nadB I03 pncA278::TnlQdCm Lab collection "'nadB 103 TnlQdCm pnuD26J ::MudJ This study TT22246 pncB 165: :TnlQ 1'.1081 (nadA-srudy pncB 165::TnlQ '" 1 081 (pnuD261 : :TT23104 TnlQdCm nadA2 19::I'.1 052(Km' Il!!l!C aroG zbh-3652::TnlQdTc)] ydeAI37::TnlQdTc pnuP' I44 Tnl0dCm nadA219::MudJ[A1052rKmr zbhz3652::Tnl0dTc)] vdeA137::Tnl0dTc Tnl0dCm nadA219::MudJ[A1052fKmr zbhz3652::TnI0dTc)] 845::Tn!0 pncA278::Tnl0dCm MudJ[A1052(;Kmr pnuC This study aroG zbh-3652::Tnl0dTcV] zdc-845::Tn!0 16 TABLE 2.1. continued Strain Genotype Source TT23105 pncA278::TnlQdCm MudJ[~1 052(Kmr pnuC This study aroG zbh-3652::TnlQdTc)] ydeA137::TnlQdTc TT23524 pncA278::TnlQdCm MudJ[~1052(Kmr pnuC This study aroG zbh-3652::TnlQdTc)] pnuP*144 zdc-845::TnlO TT23525 pncA278::TnlQdCm nadA219::MudJ[~1 052(Kmr pnuC This study aroG zbh-3652::TnlQdTc)] zdc-845::TnlO aSGSC, Salmonella Genetic Stock Centre. 17 C u l t u r e media and supplements. The E medium of Vogel and Bonner (61), supplemented with 0.2% glucose, was used as a minimal medium. Nutrient broth (NB) (0.8%; Difco) with 0.5% NaCl was used as a rich medium. Agar (1.5%; Difco) was added to E and NB media to make solid media. Bochner plates, used for selecting Tcs mutants, were prepared as described by Maloy and Nunn (2, 34); selections were made at 42°C as suggested by Kelly Hughes. Antibiotics were included at the following concentrations in rich and rninimal media, respectively: sodium ampicillin (30 mg/1 and 15 mg/1), chloramphenicol (20 mg/1 and 5 mg/1), kanamycin sulfate (50 mg/1 and 125 mg/1), and tetracycline hydrochloride (20 mg/1 and 10 mg/1). Auxotrophic supplements were added to minimal media at final concentrations described by Davis et al. (12). The chromogenic B-galactosidase substrate, 5-bromo-4-chloro-3-indolyl-B-D-galactopyranoside (Xgal) was dissolved in N,N-dimethylformamide (20 mg/ml) and added to media at a final concentration of 20 mg/1. All antibiotics and other supplements were obtained from Sigma Chemical Co.. T r a n s d u c t i o n with phage P22. The high-frequency generalized transducing phage P22 HT105/1 int-201 was used for all transductional crosses (55). Transductional crosses were achieved by adding 0.1 ml (108 to 109 phages) of a generalized transducing lysate to 0.1 ml (2 x 108 cells) of an overnight broth culture of the recipient strain. This mixture was incubated at 37°C for 1 hr before spreading on selective medium. Transductant colonies were purified on nonselective green indicator plates and were shown to be phage-free by their sensitivity to a clear plaque mutant of P22. H5 (8). Selection for NMN-assimilating revertants of a pncA p n u C strain. All pnuD* and pnuP* mutants arose spontaneously. Overnight cultures inoculated from Culture media and supplements. The E medium of Vogel and Bonner (61), supplemented with 0.2% glucose, was used as a minimal medium. Nutrient broth (NB) (0.8%; Difco) with 0.5% NaCI was used as a rich medium. Agar (1.5%; Difco) was added to E and NB media to make solid media. Bochner plates, used for selecting Tcs mutants, were prepared as described by MaJoy and Nunn (2. 34); selections were made at 42°C as suggested by Kelly Hughes. Antibiotics were included at the following concentrations in rich and minimal media, respectively: sodium ampicillin (30 mg/l and 15 mg/l), chloramphenicol (20 mg/l and 5 mg/I), kanamycin sulfate (50 mg/I and 125 mg/l), and tetracycline hydrochloride (20 mgll and 10 mgll). Auxotrophic supplements were added to minimal media at fmal concentrations described by Davis et al. (12). The chromogenic a-galactosidase substrate, 5-bromo-4-chloro-3-indolyl-B-D-galactopyranoside (Xgal) was dissolved in N,N­dimethylfonnamide (20 mg/mI) and added to media at a final concentration of 20 mg/l. All antibiotics and other supplements were obtained from Sigma Chemical Co .. Transduction with phage P22. The high-frequency generalized transducing phage P22 HT105/1 int-201 was used for all transductional crosses (55). Transductional crosses were achieved by adding 0.1 ml ( lOll to 109 phages) of a generalized transducing lysate to 0.1 ml (2 x 108 cells) of an overnight broth culture of the recipient strain. This mixture was incubated at 37°C for 1 hr before spreading on selective medium. Transductant colonies were purified on nonselective green indicator plates and were shown to be phage-free by their sensitivity to a clear plaque mutant of P22. H5 (8). Selection for NMN-assimilating rever ta nts of a pncA rnue strain. All pnuD· and pnuP· mutants arose spontaneously. Overnight cultures inoculated from 18 single colonies of a pncA pnuC strain (TT14890) were washed twice with E medium and 0.1 ml was spread on each minimal medium containing 0.1 mM NMN. Independence was assured by saving only one mutant from each culture. Isolation and genetic mapping of TnlOdTc insertions near the pnuD gene. A pnuD* strain (TT15565) was transduced to tetracycline resistance using P22 donor phage grown on a random pool of TnlOdTc insertion mutants. The Tcr transductants were screened for the loss of the recipient NMN-growth phenotype, indicative of coinheritance of a wild-type pnuD allele with the linked TnlOdTc element. Linked TnlOdTc insertions were then transduced again into the original recipient to test the linkage between the pnuD* mutation and the TnlOdTc element. To determine the chromosomal location of the pnuD* gene, a tightly linked insertion, zff-3685::Tnl0dTc (TT15574), was first mapped by Hfr mapping technique as described by Chumley et al. (10). More accurate mapping was carried out by testing transductional linkages to known markers in the region. Isolation of pnuD* deletion mutations. Spontaneous deletions extending into the pnuD* gene from two tightly linked insertions, zff-3685::TnlOdTc (TT15577) and glyA540::TnlO (TT15585), were obtained by selecting tetracycline-sensitive derivatives of these Tcr strains on Bochner Tcs plates (2, 34). Construction of t a n d em duplications of t h e p n uD region. To test dominance of pnuD* mutant alleles, tandem duplications of the pnuD region were constructed that include the region between the guaBA genes at 54.4 min and the nadB gene at 57.8 min. These merodiploids were constructed by transductional crosses using Mud insertions as regions for homologous recombination (27, 33). single colonies of a pncA pouC strain (Tf14890) were washed twice with E medium and 0.1 ml was spread on each minimal medium containing 0.1 roM NMN. Independence was assured by saving only one mutant from each culture. Isolation and genetic mapping of TnlOdTc insertions ncar the pouD gene. A pOllD· strain (TT15565) was transduced to tetracycline resistance using P22 donor phage grown on a random pool of TnlQdTc insertion mutants. The Ter transductants were screened for the loss of the recipient NMN-growth phenotype. indicative of coinheritance of a wild-type pOllD allele with the linked TnlQdTc element. Linked Tnl.QdTc insertions were then transduced again into the original recipient to test the linkage between the pouD* mutation and the TnlQdTc element. To determine the chromosomal location of the pnuO· gene, a tightly linked insertion, zfT-3685::Tn.1..QdTc (TTI5574), was first mapped by Hfr mapping technique as described by Chumley et aI. (10). More accurate mapping was carried out by testing transductional linkages to known markers in the region. holation of pnuD* deletion mutations. Spontaneous deletions e}.1ending into the pnuD* gene from two tightly linked insertions, zff-3685::Tnl.QdTc (TT I5577) and glyA540: :TnlQ (TT I5585), were obtained by selecting tetracycline-sensitive derivatives of these Tcr strains on Bochner Tcs plates (2, 34). Construction of tandem duplications of the pnuD region. To test dominance of pnuD~ mutant alleles, tandem duplications of the pnuO region were constructed that include the region between the guaBA genes at 54.4 min and the nadB gene at 57.8 min. These merodiploids were constructed by transductional crosses using Mud insertions as regions for homologous recombination (27, 33). Recombination between two transduced fragments, one carrying a guaBA::MudA (Lac~) insertion and the other a nadB::MudJ (Lac+ ) insertion, generated a duplication joint point (Lac~ Km1 ) , which can recombine with the chromosome to generate a duplication with no associated auxotrophy. This duplication was constructed in a strain (TT15589) carrying a nadA-pnuC deletion and a pncA insertion; both mutations are unlinked to the pnuD region. Isolation and physical mapping of M u d J insertions in t h e pnuD gene. Donor phage P22 grown on a random pool of MudJ insertion mutants was used to transduce a pnuD-glyA deletion strain (TT20189; isolated as described the above) to kanamycin resistance (26, 28). The Kmr transductants were screened for those with a Gly+ phenotype, indicative of coinheritance of a MudJ element and a wild-type glyA allele. Linked MudJ insertions were then transduced into a pnuD* glyA::TnlO strain (TT15585) to test the linkage between these markers and the MudJ element. Insertion joint points were sequenced to identify insertions in the pnuD gene. Five mutants with MudJ inserted in the pnuD region (TT20190 to 20194) were isolated and physically mapped or sequenced by PCR as follows. Primers were designed corresponding to the region downstream of the glyA gene (Its sequence was previously reported) (56), TP533 (5'-GCGCTTGTGTTGTGAAAATG-3') and to the left and right ends of the Mu transposon, TP241 (5'-CCCGAATAATCCAATGTCCTCCCGG-3') and TP81 (5'-GAAACGCTTTCGCGTTTTTCGTGCG-3'), respectively. For each MudJ insertion, two PCR reactions were set up, each with the genomic primer at one side and one of the two MudJ primers at the other side. 19 Recombination between two transduced fragments, one carrying a guaBA::MudA (Lac-) insertion and the other a nadB::Mudl (Lac+) insertion, generated a duplication joint point (Lac- Kmf), which can recombine with the chromosome to generate a duplication with no associated auxotrophy. This duplication was constructed in a strain (TTI5589) carrying a nadA-pnuC deletion and a pncA insertion; both mutations are unlinked to the pnuD region. Isolation and physical mapping of MudJ insertions in the pnuD gene. Donor phage P22 grown on a random pool of Mudl insertion mutants was used to transduce a pnuD-glyA deletion strain (TT20189; isolated as described the above) to kanamycin resistance (26, 28). The Kmf transductants were screened for those with a Gly+ phenotype, indicative of coinheritance of a Mud] element and a wild-type g.lyA allele. Linked Mud] insertions were then transduced into a pnuD* g.lyA::Tnl0 strain (TTI5585) to test the linkage between these markers and the Mudl element. Insertion joint points were sequenced to identify insertions in the pnuD gene. Five mutants with Mud] inserted in the pnuD region (TT20190 to 20194) were isolated and physically mapped or sequenced by PCR as follows. Primers were designed corresponding to the region downstream of the g.lyA gene (Its sequence was previously reported) (56), TP533 (5'-GCGCTTGTGTTGTGAAAATG-3') and to the left and right ends of the Mu transposon, TP241 (5'-CCCGAATAATCCAATGTCCTCCCGG-3') and TP81 (5'-GAAACGCTTTCGCGTTTTTCGTGCG-3'), respectively. For each Mud] insertion, two PCR reactions were set up, each with the genomic primer at one side and one of the two Mud] primers at the other side. As expected, only one reaction mixture yielded a product depending on the orientation of the MudJ element. To confirm the orientation, the five strains (TT20190 to 20194) were plated on Xgal plates; ]ac expression led to blue color suggested that the orientation allowed transcription of the inserted ]ac operon. The approximate location of each insertion was deduced from the length of the PCR product; the precise nucleotide positions of three insertions were determined by sequencing. Isolation and mapping of TnlOdTc insertions near the pnuP gene. Mutants with an insertion of TnlOdTc near the pnuP gene were isolated by the same strategy as that used to obtain TnlOdTc insertions near the pnuD gene. A pnuP* mutant (TT15573) was transduced to tetracycline resistance using P22 donor phage grown on a random pool of TnlOdTc insertion mutants. The Tcr transductants were screened for the loss of the recipient NMN-growth phenotype. Linked TnlOdTc insertions were then transduced again into the original recipient to determine the linkage between the pnuP* mutation and the TnlOdTc element. To map the pnuP* mutation, the flanking sequence of a tightly linked insertion, zxx-9180::Tnl0dTc (TT23104), was determined by using a single oligonucleotide primer TP328 (5'-TATGCCGCCATTATTACGAC-3') that extends out of TnljOdTc element. Nucleotide sequencing of the pnuD and pnuP regions. A 2,147-bp DNA fragment including the entire pnuD gene was amplified by PCR from strain TT20194 (pnuD+ zff-3868::MudJ) using primers TP533 and TP81 (See above). Four strains (TT15564, 15565, 15572, and 15586) carrying a pnuD* or pnuD mutation were sequenced with TP533 and TP1094 (5'-CGCTCGCCGGATTTGAAGC-3*) as primers. 20 As expected, only one reaction mixture yielded a product depending on the orientation of the MudJ element. To confirm the orientation, the five strains (TT20190 to 20194) were plated on Xgal plates; lac expression led to blue color suggested that the orientation allowed transcription of the inserted lac operon. The approximate location of each insertion was deduced from the length of the PCR product; the precise nucleotide positions of three insertions were determined by sequencing. Isolation and mapping of TnlOdTc insertions near the pnuP gene. Mutants with an insertion of Tnl0dTc near the pnuP gene were isolated by the same strategy as that used to obtain TnlQdTc insertions near the pnuD gene. A pnuP* mutant (TTI5573) was transduced to tetracycline resistance using P22 donor phage grown on a random pool of TnlQdTc insertion mutants. The Tcr transductants were screened for the loss of the recipient NMN-growth phenotype. Linked Tnl0dTc insertions were then transduced again into the original recipient to determine the linkage between the pnuP* mutation and the TnlQdTc element. To map the pnuP* mutation, the flanking sequence of a tightly linked insertion, zxx-9180::TnlOdTc (TT231 04), was determined by using a single oligonucleotide primer TP328 (5'-TATGCCGCCATTATTACGAC-3') that extends out of Tnl0dTc element. Nucleotide sequencing of the pnuD and pnuP regions. A 2,147-bp DNA fragment including the entire pnuD gene was amplified by PCR from strain TT20194 (pnuD+ zff-3868::MudJ) using primers TP533 and TP81 (See above). Four strains (TTI5564, 15565, 15572, and 15586) carrying a pnuD* or pnuD mutation were sequenced with TP533 and TPI094 (5'-CGCTCGCCGGATTTGAAGC-3') as primers. An 1,759-bp fragment including the entire pnuP gene was amplified from strain TT23525 (pnuP+ ) or TT23524 (pnuP*144) using primers TP1090 (5'- AGTATGCCAGGACATCATCAG-3') and TP1091 (5*-TCATCAGTTCCGCTTCCA C-3'). Primer TP 1090 was designed as inferred from the Salmonella genome sequence and primer TP 1091 was inferred from the previously reported sequence of the mar operon (57). By the same strategy, an overlapping 1,913-bp fragment including the remainder of the mar locus was amplified using primers TP1092 (5'-TTCAATCCCCCCTTTACCC- 3') and TP1093 ( 5 A G AC A AACGCT AATTTC AGACC-3'). The DNA template was extracted with the QIAquick Gel Extraction Kit (Qiagen) from agarose gels in TAE (Tris-acetate/EDTA) buffer and precipitated with ethanol as described by Sambrook et al. (54). Custom-made oligonucleotides used as PCR primers and in DNA sequencing were prepared by Robert Schackmann of the DNA and Peptide Facility at the University of Utah. DNA sequencing was performed in the same facility using the automated fluorescent sequencing coupled with PCR. Computer analysis of sequence. Sequences were analyzed with the Genetics Computer Group programs (version 7; Madison, WI) (14). Sequence homologies were searched with PSI-BLAST program (1). The DNA Strider program (version 1.2) (35) was used for hydrophobicity plots according to the method of Kyte and Doolittle (31). Nucleotide sequence accession number. The 2,129-bp fragment containing the pnuD gene was submitted to GenBank and assigned to the accession number AF306512. The 1,759-bp fragment containing the pnuP gene was submitted to GenBank and assigned to the accession number AF440748. 21 An 1,759-bp fragment including the entire pnuP gene was amplified from strain TT23525 (pnuP' ) or TT23524 (pnuP*144) USillg pnmers TPI090 (5'­AGTATGCCAGGACATCATCAG- 3') and TPI091 (5'-TCATCAGTTCCGCTTCCA C-3'). Primer TPI090 was designed as inferred from the Salmonella genome sequence and primer TP J 091 was inferred from the previously reported sequence of the mar operon (57). By the same strategy. an overlapping l ,913-bp fragment including the remainder of the mar locus was amplified using primers TPI092 (5'-TTCAATCCCCCCTTTACCC- 3') and TPI093 (5'-AGACAAACGCTAATTTCAGACC-3'). The DNA template was extracted with the QlAquick Gel Extraction Kit (Qiagen) from agarose gels in TAE (Tris-acetateIEDTA) buffer and precipitated with ethanol as described by Sarnbrook et aI. (54). Custom-made oligonucleotides used as peR primers and in DNA sequencing were prepared by Robert Schackmann of the DNA and Peptide Facility at the University of Utah. DNA sequencing was perfonned in the same facility using the automated fluorescent sequencing coupled with peR. Computer analysis of sequence. Sequences were analyzed with the Genetics Computer Group programs (version 7; Madison, WI) ( 14). Sequence homologies were searched with PSI-BLAST program (I). The DNA Strider program (version 1.2) (35) was used for hydrophobicity plots according to the method of Kyte and Doolittle (3 I). Nucleotide sequence accession number. The 2, 129-bp fragment containing the pnuD gene was submitted to GenBank and assigned to the accession number AF3065 12. The 1,759-bp fragment containing the pnuP gene was submitted to GenBank and assigned to the accession number AF440748. Isolation of PnuD*-inactivating mutations. Generalized mutagenesis with diethylsulfate (DES) was done as described (50). A saturated DES solution was made by adding 0.2 ml of DES to 5 ml of E medium with no carbon source and incubating for 1 hr at 37°C. Overnight culture of a pnuP* strain (TT15565) was diluted 50-fold into the DES solution and mutagenized for 30 min at 37°C. An aliquot was removed, diluted 50- fold into fresh medium without mutagen, and grown overnight to saturation. This culture was diluted 106-fold and a 0.1 ml aliquot was plated on an NB plate. After overnight incubation at 37°C, colonies were replica-plated onto minimal media with and without 0.1 mM NMN to identify mutants that could not use NMN as a pyridine source. To seek p n u P mutations that might block the PnuC-dependent route of NMN assmilation, the p n u P region was mutagenized by the method of Hong and Ames (12, 23). Transducing phage P22 grown on a glyA::TnlO strain (TT418) was mutagenized with hydroxylamirie and this mutagenized phage stock was used to transduce a nadB pncA strain (TT12990) to tetracycline resistance and transductants were tested for growth on NMN. fi-Galactosidase assay. The specific activity of B-Galactosidase was assayed in cells permeabilized by sodium dodecyl sulfate and chloroform determined as described by Miller (39). Assays were performed at least twice with samples from independent cultures. The activity is expressed as the following units: [(A4 2o 1-75 x A5 5o)/(A6oo x reaction time x volume)] x (1 nM/0.0045 ml cm) x 1.7 ml. assimilating pnuC-nadA pnuP of assimilation. pnuC p n u P 22 Isolation of PnuD*-inacth'ating mutations. Generalized mutagenesis witb diethylsulfate (DES) was done as described (50). A saturated DES solution was made by adding 0.2 ml cfDES 10 5 m1 ofE medium with no carbon source and incubating for 1 hr at 3rc. Overnight culture of a pnllD* strain (IT I5565) was diluted 50-fold into the DES solution and mutagenized for 30 min at 37°C. An aliquot was removed, diluted 50- fold into fresh medium without mutagen, and grown overnight to saturation. This culture was diluted 106.fold and a 0,1 ml aliquot was plated on an NB plate. After overnight incubation at 37°C. colonies were replica-plated onto minimal media with and without 0.1 mM NMN to identify mutants that could not use NMN as a pyridine source. To seek pnuO mutations that might block the Poue -dependent route of NMN assmilatiol1, the pnuD region was mutagenized by the method of Hong and Ames (12, 23). Transducing phage P22 grown on a we.::TnlO strain (TT4IS) was mutagenized with hydroxylamine and this mutagenized phage stock was used to transduce a nadB pncA strain (TT12990) to tetracycline resistance and transductants were tested for growth on NMN. Il-Galactosidase assay. The specific activity of B-Galactosidase was assayed in cells permeabilized by sodium dodecyl sulfate and chloroform detennined as described by Miller (39). Assays were performed at least twice with samples from independent cultures. The activity is expressed as the following units: [(A42o - 1.75 x Asso)/(A60o x reaction time x volume)] x (1 nMJO.0045 ml em) x I.7 ml. Selection for NMN-assimilating mutants of a pncA pUllC-DadA puuD strain. Spontaneous mutants were selected in this strain, which lacked all routes of NMN assimi lation. Overnight cultures of a pncA ~ pnuD strain (TT20190) were washed twice with E medium and 0.1 ml was spread on each minimal medium containing 0.1 mM or 0.3 mM NMN. Colonlies that arose were picked and analyzed. p n u C 108 - further. NMN 0.1 mMa mM ACnadA ACnadA p n uO - - ACnadA p n u O pnuD*ACnadA ACnadA p n u O - ACnadA p n u O - 23 washed twice with E medium and 0.1 m1 was spread on each minimal medium containing 0.1 mM or 0.3 mM NMN. Colonlies that arose were picked and analyzed. Results Isolation of suppressors of a pncA pnuC double mutant. A strain containing a pncA insertion and a nadA-pnuC deletion (TT14890) is unable to synthesize NAD or assimilate NMN to satisfy its pyridine requirement, because it lacks ability to used either Nm or NmR, which are derived from NMN in the periplasm. Overnight cultures of this strain were washed and plated on the minimal medium plates containing 0.1 mM NMN. After 48 hrs, NMN-growing mutant colonies arose at a frequency of about 1 per 108 _ plated cells. Ten independent revertants (TT15564 to 15573) were examined further. Phenotypes of four of these suppressor mutants are shown in Table 2.2. TABLE 2.2. Phenotypes of pncA pnuC suppressor mutants on NMN Growth on NMN Strain Relevant genotype 0.3 mM TT15148 ll(nadA pnuC) + + TT12990 nadB pncA + + TT14890 pncA ll(nadA pnuC) TT15564 pncA ll(nadA pnuC) pnuD* 135 + + TT15565 pncA ll(nadA pnuC) pnuD* 136 + + TT15572 pncA ll(nadA pnuC) pnuD*143 + TT15573 pncA ll(nadA pnuC) pnuP*144 + aStrains TT15572 and 15573 did not grow on 0.1 mM NMN overnight incubation, but started to grow after 24 hrs. 24 Genetic mapping of the p n u P region. To facilitate mapping, a TniOdTc insertion was identified that lies near one of the new mutations, pnuP* 136(TT15565) (Materials and Methods). This insertion fzff-3685::Tn!0dTc in TT15577) was 43.5% cotransducible with pnuD*136 and 30 to 45% cotransducible with all other mutations except pnuP*144. We have designated the gene linked to this TniOdTc element p n uP (TT15564 to 15572) and the unlinked mutation pnuP (TT15573). The TniOdTc insertion near p n u P * was first localized to the region between 54.4 and 57.8 min by Hfr mapping by the method of Chumley et al. (10). Transduction tests with several genetic markers in this region revealed 41.5% linkage to glvMO (TT21997) (Fig. 2.1); the pnuD*136 mutation was 85%) cotransducible with an insertion glyA540::TnlO (TT418) in a region known to include the pepB and purG loci. Results of these crosses indicated that the pnuP* gene is located counterclockwise of the glyA locus at 55.9 min, unlinked to any known genes of NAD metabolism. Deletion mutations generated by TnlO in the gene. Spontaneous pnuP deletion mutants were isolated from two pnuP* mutants (TT15577 and TT15585) as described in Materials and Methods. First, two independent NMN" Gly+ deletions (A1879 in TT20186) were isolated as Tcs revertants of a zff-3685::Tn!0dTc pnuD*136 strain (TT15577). These deletions demonstrated that the p n u P locus can be deleted without deleterious effects. Second, 13 NMN" Gly" deletions (A1890 in TT20189) and one N M N + Gly" Pur" deletion (A1904 in TT20180) were isolated as Tcs revertants of a TnlO order Genetic mapping of the pnuD region. To facilitate mapping, a Tn10dTc insertion was identified that lies near one of the new mutations, pnuD* 136(TT15565) (Materials and Methods). This insertion (zff-3685::Tn10dTc in TT15577) was 43.5% cotransducible with pnuD* 136 and 30 to 45% cotransducible with all other mutations except pnuP*144. We have designated the gene linked to this Tn10dTc element pnuD (TT15564 to 15572) and the unlinked mutation pnuP (TT15573). The Tn10dTc insertion near pnuD* was first localized to the region between 54.4 and 57.8 min by Hfr mapping by the method of Chumley et al. (l0). Transduction tests with several genetic markers in this region revealed 41.5% linkage to glyA 10 (TT21997) (Fig. 2.1); the pnuD* 136 mutation was 85% cotransducible with an insertion glyA540::Tn10 (TT418) in a region known to include the ~ and purG loci. Results of these crosses indicated that the pnuD* gene is located counterclockwise of the glyA locus at 55.9 min, unlinked to any known genes ofNAD metabolism. Deletion mutations generated by TnlO in the gene. Spontaneous pnuD deletion mutants were isolated from two pnuD* mutants (TT15577 and TT15585) as described in Materials and Methods. First, two independent NMN- Gly+ deletions (~1879 in TT20186) were isolated as Tcs revertants of a zff-3685::Tn10dTc pnuD*136 strain (TT15577). These deletions demonstrated that the pnuD locus can be deleted without deleterious effects. Second, 13 NMN- Gly- deletions (~1890 in TT20189) and one NMN+ Gly- Pur- deletion (~1904 in TT20180) were isolated as Tcs revertants of a pnuD*136 glyA540::Tn10 strain (TT15585). These results confirmed the gene order pnuD-glyA-purG and demonstrated that nothing between the glyA and purG genes is essential for life or NMN transport. These deletions are indicated in Fig. 2.1. Other loss- 136 g l y A 5 40 54.4 min ^801 z£3685 \g l y A ] J ^ 7 5 6 > 0 m i n 65% 90% 43.5% 41.5% 5% 85% 14% 1 5Z2Z2 Tn76>dTc TnlO 1 A1890 1 purG 1739 r Z 7 57.8 min W nadB pnuP pnuP glyA gene d e t e r m i n e d P 2 2 - m e d i a t e d t r a n s d u c t i o n a l c r o s s e s s c o r i n g 200 extent of the three deletions is indicated just below the map. 25 pnuD* 136 glyA540 purGl739 iff-80l zjf-3685 T 57.8 min guaAB t pepB t nadB ~~~~~~---C~~~--~~~ 12% 4% 41879 ---I M890 41904 TnlOdTc FIG. 2.1. Genetic linkages of the pnuD region in Salmonella serovar Typhimurium. The pnuD gene was 90% linked to the ~gene at 55.9 min. All the linkages were determined by P22-mediated transductional crosses by scoring 100 or 200 transductants. The arrowheads point to the donor marker selected in the cross. The of-function mutations in p n u P (insertions and point mutations) will be described later. Dominance test of the p n u P * mutations. If the p n u P * mutations allow N MN transport by structural alteration of an transporter for some other compound, they might be expected to be dominant. To test dominance, the pnuP* allele was introduced into a strain (TT15589) with a tandem duplication of a chromosomal segment from guaBA (54.4 min) to nadB (57.8 min) including the p n u D + gene. This strain carries an unlinked nadA-pnuC deletion and an unlinked pncA insertion, and thus lacks both standard routes of NMN assimilation. A pnuP* mutation was selectively introduced into the duplication recipient by transduction using donor phage grown on a strain (TT15585), which carries a pnuD*136 mutation and a 85%-linked insertion, glyA54Q::TnlO. Of the Tcr transductants, 85% were able to grow on 0.1 mM NMN. Segregants that lost the duplication included both NMN+ and NMN" types demonstrating that the NMN+ duplication strain (TT15590) carried both pnuP* and p n u D + alleles and that the p n u P* allele is dominant. Isolation and physical location of MudJ insertions in the p n u P gene. Insertions of MudJ (Lac+ ) near the glyA gene were isolated and genetically mapped, and the five closest to pnuP* were located by PCR and sequenced as described in Materials and Methods. Four insertions (pnuD261 to 264. in TT20190 to 20193, respectively) were within the p n u P coding sequence, while the fifth (zff-3868 in TT20194) was in the immediately adjacent csiE homologue and redefined as csiEl::MudJ. The pnuD264::MudJ insertion was in the overlapped region of the p n u P and csiE genes. The csiEl::MudJ insertion was used further for sequencing the p n u P gene. 26 of-function mutations in pnuD (insertions and point mutations) will be described later. Dominance test of the pnuD* mutations. If the pnuD* mutations allow NMN transport by structural alteration of an transporter for some other compound, they might be expected to be dominant. To test dominance, the pnuD* allele was introduced into a strain (TTI5589) with a tandem duplication ofa chromosomal segment from guaBA (54.4 min) to nadB (57.8 min) including the pnuD+ gene. This strain carries an unlinked nadA­pnuC deletion and an unlinked pncA insertion, and thus lacks both standard routes of NMN assimilation. A pnuD* mutation was selectively introduced into the duplication recipient by transduction using donor phage grown on a strain (TT15585), which carries a pnuD*136 mutation and a 85%-linked insertion, glyA540::Tn10. Of the Tcr transductants, 85% were able to grow on 0.1 mM NMN. Segregants that lost the duplication included both NMN+ and NMN- types demonstrating that the NMN+ duplication strain (TTI5590) carried both pnuD* and pnuD+ alleles and that the pnuD* allele is dominant. Isolation and physical location of MudJ insertions in the pouD gene. Insertions of MudJ (Lac +) near the g.lyA gene were isolated and genetically mapped, and the five closest to pnuD* were located by PCR and sequenced as described in Materials and Methods. Four insertions (pnuD261 to 264, in TT20190 to 20193, respectively) were within the pnuD coding sequence, while the fifth (zff-3868 in TT20194) was in the immediately adjacent csiE homologue and redefined as csiE1 ::MudJ. The pnuD264::MudJ insertion was in the overlapped region of the pnuD and csiE genes. The csiEI ::MudJ insertion was used further for sequencing the pnuD gene. Two of the identified Mud-lac fusions within the p n u P gene (pnuD261::MudJ and pnuD262::MudJ) were in the correct orientation for expression by the pnuP promoter and were used to study regulation of the p n u P gene. Physical location of these five MudJ insertions is depicted in Fig. 2.2. Nucleotide sequence analysis of the pnuP region. The DNA sequence determined as described in Materials and Methods included the entire p n u P gene and part of a second downstream open reading frame, the csiE gene (Fig. 2.3). The complete p n uP gene is inferred to encode a 379-amino-acid protein. The most probable transcription start site was located 51 bp upstream of the start codon. A putative promoter site for transcription was identified by the sequence TTGAAC-22bp-TATGTT, which provides a good match to the consensus sequence (38). A putative ribosome binding site, GGAGG, was found at nucleotides -10 to - 6 . An in-frame TGA stop codon occurred at nucleotides 1138 to 1140. The most probable transcription terminator was found at the site, which includes a G-C-rich inverted repeat sequence, GCGC-6 bp-GCGC, followed by a run of 5U's and ends 141 bp downstream of the stop codon. The 3' end of the convergent csiE transcript gene overlaps the 3' end of the p n u P coding sequence for 119 bp. The csiE gene is known as a stationary-phase-inducible gene under the control of sigma S and the cAMP-CRP complex in E. coH (37, 63). The P n uD protein is an MFS permease. The deduced amino acid sequence of the PnuD protein showed a high level of similarity to a few members of the MFS proteins. The length of the PnuD protein, 379 amino acids, was close to the length range of MFS, about 400 to 600 amino acids. The hydrophobicity profile of the PnuD amino 27 Two of the identified Mud-lac fusions within the pnuD gene (pnuD261: :MudJ and pnuD262::MudJ) were in the correct orientation for expression by the pnuD promoter and were used to study regulation of the pnuD gene. Physical location of these five MudJ insertions is depicted in Fig. 2.2. Nucleotide sequence analysis of the pnuD region. The DNA sequence determined as described in Materials and Methods included the entire pnuD gene and part of a second downstream open reading frame, the csiE gene (Fig. 2.3). The complete pnuD gene is inferred to encode a 379-amino-acid protein. The most probable transcription start site was located 51 bp upstream of the start codon. A putative promoter site for transcription was identified by the sequence TTGAAC-22bp-TATGTT, which provides a good match to the consensus sequence (38). A putative ribosome binding site, GGAGG, was found at nucleotides -10 to -6. An in-frame TGA stop codon occurred at nucleotides 1138 to 1140. The most probable transcription terminator was found at the site, which includes a GeC-rich inverted repeat sequence, GCGC-6 bp-GCGC, followed by a run of 5U's and ends 141 bp downstream of the stop codon. The 3' end of the convergent csiE transcript gene overlaps the 3' end of the pnuD coding sequence for 119 bp. The csiE gene is known as a stationary-phase-inducible gene under the control of sigma S and the cAMP-CRP complex in E. coli (37, 63). The PnuD protein is an MFS permease. The deduced amino acid sequence of the PnuD protein showed a high level of similarity to a few members of the MFS proteins. The length ofthe PnuD protein, 379 amino acids, was close to the length range of MFS, about 400 to 600 amino acids. The hydrophobicity profile of the PnuD amino csiEl pnuD264 pnuD263 pnuD262 pnuD261 w v csiE pnuP UK pnuP determined MudJ. MudJ. csiEl::MudJ. .c.s..i.E. l pnuD264 pnuD263 pnuD262 ~~ ....... V VVV pn.u..D..2. 6l V csiE I ~nuD ---------------------~~ .-~~----------------- 28 glyA FIG. 2.2. Relative location of MudJ insertions in the pnuD locus. The approximate position and the orientation of each insertion were detennined via PCR. The exact nucleotide posi­tion of three insertions, pnuD261::MudJ, pnuD264::MudJ, and csiE1::MudJ, was deter­mined by sequencing; they were located between + 77T and + 78C, + 1078A and + 1079T, and +1919C and +1920, respectively (FIG. 2.3). The arrow above the triangle indicates the direction of lacZ expression. The arrow below each gene indicates the transcription direc­tion. pnuP ATG ( • ) , putative promoter region (-35 and - 1 0 ) and a putative ribosome binding site (RBS) as gray boxes. Another boxed sequence represents PUR box (See Discussion). Two black boxes indicate the sites of missense mutations causing pnuP* phenotype (aa52) and loss of the NMN assimilation activity of the pnuP* mutants (aal33), respectively (Table 2.3). Stop codons are designated by asterisks. The most probable transcription terminator is underlined. Sequences of the primers TP533 and TP 1094, used for sequencing the mutant strains, are underlined with arrows. Another arrow indicates the transcriptional direction of the csiE gene, overlapped with the p n u P coding sequence. The sequence shown was submitted to GenBank with accession number AF306512. 29 FIG. 2.3. Nucleotide sequence and deduced amino acid sequence of the Salmonella serovar Typhimurium pnuD gene. The deduced amino acids are represented below the first nucleotide of each codon in the coding sequence. Nucleotide + 1 denotes the A of the A TG of the initiator methionine. Residues preceding it are indicated by negative numbers. The most probable transcription start site is indicated as the symbol (.), together with a 10) pnuD* pnuD* aa133), TP1094, pnuD - 210 CCGGTTTACGCATAAGCGCTTGTGTTGTGAAAATGGCCCGATGGCGCTGCGTCTGTCGGGCATTACTTATGCGCCATAACGTGACGCCGC P V Y A * primer TP533 > glyA end -35 -10 • PUR Box -12 0 CGGAAGTCATTATCCGGCGGCGTTGCCGTTfGAACTTCTCCCTCCGCCTGTTAACCTyATGTTACCCTTGCCA|GCGCAACCGTTTACCG|C RBS +1 -30 CAGACTATCGCCTCCAAACGGGAGGGAATCATGGCATTGCATTCCACGCGCTGGCTGGCGCTCAGTTATTTCACCTACTTCTTTAGTTAC pnuD start M A L H S T R W L A L S Y F T Y F F S Y GGTATTTTTCTGCCCTTCTGGAGCGTCTGGCTCAAAGGTCTTGGGCTAACGCCGGAAACCATCGGTCTTCTGCTGGGCGTGGGTCTGGTC G I F L P F W S V W L K G L G L T P E T I G L L L G V G L V 151 GCG^JTTTCTCGGTAGTCTGCTCATTGCGCCTCGCGTAAGCGATCCTTCGCGGTTGATCTCCGCGCTGCGCGTCCTGGCATTGCTGACG A G F L G S L L I A P R V S D P S R L I S A L R V L A L L T 241 CTGGTATTTGCGCTGGCATTTTGGGCAGGAACGCATGTCGCGTGGCTAATGGTGGTGATGGTTGGGTTTAACCTCTTCTTTTCGCCACTG L V F A L A F W A G T H V A W L M V V M V G F N L F F S P L 331 GTGCCGCTGACCGATGCGCTGGCCAATACCTGGCAAAAGCAAATTACCCTGGACTATGGTCGGGTGFFLGGCTGTGGGGGTCCATCGCCTTC V P L T D A L A N T W Q K Q I T L D Y G R V G L W G S I A F 421 GTGATAGGGTCGGCGCTGACCGGTAAGCTGGTGAGTTTATACGATTACCAGGCGATTCTGGCGCTACTGACGCTCGGCGTCGCCTCGATG V I G S A L T G K L V S L Y D Y Q A I L A L L T L G V A S M 511 TTGCTGGGTATGTTGTTACGTCCCAGCGTGCCGCCGCAAGGGGAAAGCCGTCAGCAGGAGAGCGCAGGCTGGCCCGCCTGGCGTACGCTG L L G M L L R P S V P P Q G E S R Q Q E S A G W P A W R T L 601 GTGGCACAAAGTTGGCGTTTTCTCGCCTGCGTCTGTTTGCTGCAAGGGGCGCACGCCGCCTACTACGGTTTTAGCGCCATCTACTGGCAG V A Q S W R F L A C V C L L Q G A H A A Y Y G F S A I Y WQ 691 GGGGCTGGCTATTCAGCATCGGCGGTCGGCTATTTATGGTCGTTGGGCGTGGTGGCGGAAGTGATCATTTTCGCCCTGAGTAAAAAGCTA G A G Y S A S A V G Y L W S L G V V A E V I I F A L S K K L 781 TTCCGCCGGTTTAGCGCCCGCGACTTGCTGTTGCTTTCCGCTGTGTGCGGCGTGGTACGCTGGGGACTGATGGGCTGGAGTACGGCGTTG F R R F S A R D L L L L S A V C G V V R W G L M G W S T A L 871 CCGTGGCTGATTGTGATACAAATCCTGCACTGTGGCACGTTTACCGTTTGCCATCTGGCCGCCATGCGCTATATCGCTGCGCGTCAGGGA P W L I V I Q I L H C G T F T V C H L A A M R Y I A A R Q G csiE end * < 961 AGCGAGGTGATTCGTTTGCAGGCCGTCTATTCCGCTGTGGCGATGGGCGGCAGTATCGCGATTATGACGGTATTTGCCGGTTTTCTGTAT S E V I R L Q A V Y S A V A M G G S I A I M T V F A G F L Y 1051 CAACATCTGGGCGGCGGCGTATTCTGGATTATGGCGCTGGTGGCATTGCCCGCTATCTTCCTTCGGCCCAAAGTGGTTGCCGCGTCATGA Q H L G G G V F W I M A L V A L P A I F L R P K V V A A S * pnuD end 1141 TTCCAGGATCTTGCGGATCTGCTGCTGTTGGTGCTCTGTGAGGGCGCGATCCGCATGAATTAACGGCGGTGAGAAGAGCGGTAGCGGCGT 1231 GGCGTAAGGGGTAACAATCAGCGCTACGCCGCGCGGGCATCCCTCTTTTTGAAACGCCTGGAGCGATATCCGTCTGATATTCAGCGGCAA transcription terminator 1321 CAGGGTCAGTTCACGCAGTTGCTGCTCAATATGCGTCTCCAGCGCATCATTTTTATCCGCTAACAATACGATCTGTCTCTCATGCAAATC 1411 GTTATCCTGCATTAGCCAGGCGCCGAAAATCACTGCCACCAGCCCCGTCTCTTCCTCGGAAAAGTGGATACCATATTCGGCTTCAAATCC < primer 1501 GGCGAGCGCCTCACGTGTGGTGCGAACCAGACGCGGATAAAGACGGTTAAACTCTTCCGGAAGCGTATTATCGATGCCAATCGTGAATAA TP1094 15 91 ACTTCGGTTCAGCGCCTGGGCGAGATGGACATACAGCTGATCGTTAAGCCCTTGTTCGTCGCTGAAGCGTACGTTCCCTTTTTCTCGAAA 1681 GCGCAGCACCAGCCGCGCCACCTCCAGTCGCAGTTGCTGCGCCCGTTGATGCGTGTCGCGAATCGGGTCGGGGAGGCGAATCATTGAAAA 1771 CAGTAACGCCATAAACAGCGCTTCATTGAGCGGTGCGGCCTGCATCACGTGGCGCTGCCAGTGACGACCAATTTCCTGGGCGAGTGGATA 1861 TTCCGCGCACGACTGCGCCCAAATCTGCTGCACAGGGTTGAATTCCGGGGTGATGCGCC 30 CCGGTTTACGCATAAGCGCTTGTGTTGTGAAAATGGCCCGATGGCGCtGCGtCTGTCGGGCATTACTTATGCGCCAtAACGTGACGCCGC P V Y II primer TP533 --------t glyA end -35 -10 . PUR Box -120 CGGAAGTCATTATCCGGCGGCGTTGCCG TTCTCCCTCCGCCTGTTAACCTiATGTfACCCTTGCCAFCGCAACCGTTTACCgc -30 RBS +1 CAGACTATCGCCTCCAAAC~GAATCATGGCATTGCATTCCACGCGCTGGCTGGCGCTCAGtTATTT CACCTACTTCTTTAGTTAC pnuD start MAL H S TRW L A L S y r T y r r s y 61 GGTATTTTTCTGCCCTTCTGGAGCGTCtGGCtCAAAGGtCTtGGGCTAACGCCGGAAACCATCGGTCTTCTGCTGGGCGTGGGtCTGGTC GIFLPrWSVWLKGLGLTPETIGLLLGVGLV GCGillTTTCTCGGTAGTCTGCTCATTGCGCCTCGCGTAAGCGATCCTTCGCGGTTGATCTCCGCGCTGCGCGtCCtGGCATTGCtGACG !i1 r LGSLLIAPRVSDPSRLISALRVLALLT CTGGTATTtGCGCTGGCATTTTGGGCAGGAACGCATGTCGCGTGGCTAATGGTGGTGATGGTTGGGTTTAACCTCTTCtTTTCGCCACTG LVFALAFWAGTHVAWLMVVMVGFNLFFSPL 331 GTGCCGCTGACCGATGCGCTGGCCAATACCTGGCAAAAGCAAATTACCCTGGACTATGGTCGGGTGlllCTGTGGGGGTCCATCGCCTTC V P LTD II L 1\ N T \Ii Q K 0 1 T L D Y G R V ~ L W G S I A F 421 GTGATAGGGTCGGCGCTGACCGGTAAGCTGGTGAGTTTATACGATTACCAGGCGATTCTGGCGCTI\CTGACGCTCGGCGTCGCCTCGI\TG 1 GSA LTG K L V SLY D Y Q A I L ALL T L G V A S M TTGCTGGGTI\TGTTGTTACGTCCCAGCGTGCCGCCGCAAGGGGAAAGCCGTCAGCAGGAGAGCGCAGGCTGGCCCGCCTGGCGTACGCTG G M L R P S V P P 0 G E S ROO E SAG W P 1\ W R T L GTGGCACAAAGTTGGCGTTTTCTCGCCTGCGTCTGTTTGCTGCAAGGGGCGCACGCCGCCTACTACGGTTTTAGCGCCATCTACTGGCAG 0 \Ii F L 1\ C V C L LOG A H A A Y Y G F SAl Y W 0 GGGGCTGGCTATTCAGCATCGGCGGTCGGCTATTTATGGtCGtTGGGCGtGGTGGCGGAAGTGATCATTTTCGCCCTGAGTAAAAAGCTA GAGYSASAVGYLWSLGVVAEVllfALSKKL TTCCGCCGGTTTAGCGCCCGCGACTtGCTGTTGCTTTCCGCTGTGTGCGGCGTGGTACGCTGGGGACTGATGGGCTGGAGTACGGCGtTG fR RfSARDL LL LSAVCGVVRWG LMG W STA L CCGTGGCTGATTGTGATACAAATCCTGCACTGTGGCACGTTTACCGTTTGCCATCTGGCCGCCATGCGCTATATCGCTGCGCGTCAGGGA 1 V I 0 I L H C G T F T V C H L 1\ A M R Y I A A 'R 0 G csiE end ~,--- 961 I\GCGAGGTGATTCGTTTGCAGGCCGTCTATTCCGCTGTGGCGATGGGCGGCAGTATCGCGATTATGACGGTATTTGCCGGTTTTCTGTAT SEVIRLQAVYSAVAMGGSIAIMTVfAGFLY 1051 CAACATCTGGGCGGCGGCGTATTCTGGATTATGGCGCTGGTGGCATTGCCCGCTATCTTCCTTCGGCCCAAAGTGGTTGCCGCGTCATGA OHLGGGVFWI M ALVALPAIFLRPKVVAAS pnuD end: 11.1 TTCCAGGATCTTGCGGATCTGCTGCTGTTGGTGCTCTGTGAGGGCGCGATCCGCATGAATTAACGGCGGTGAGAAGAGCGGTAGCGGCGT 1231 GGCGTAAGGGGTAACAATCI\GCGCTACGCCGCGCGGGCI\TCCCTCTTTTTGAAACGCCTGGAGCGATATCCGTCTGATATTCAGCGGCAA ----transcription t e~ina tor--- 1321 CAGGGTCAGTTCACGCAGTTGCTGCTCAATATGCGTCTCCAGCGCATCATTTTTATCCGCTAACAATACGATCTGTCTCTCATGCAAATC 1411 GTTATCCTGCATTAGCCAGGCGCCGAAAATCACTGCCACCAGCCCCGTCTCTTCCTCGGAAAAGTGGATACCATATTCGGCTTCAAATCC _ 1501 GGCGAGCGCCTCI\CGTGTGGTGCGAACCAGACGCGGATAAAGACGGTTAAACTCTTCCGGAJI.GCGTATTATCGATGCCAATCGTGAATAA TP I 094 1591 ACTTCGGTTCAGCGCCTGGGCGAGATGGACATACAGCTGATCGTTAAGCCCTTGTTCGTCGCTGAAGCGTACGTTCCCTTTTTCTCGAAA 1681 GCGCAGCACCAGCCGCGCCACCTCCAGTCGCAGTTGCTGCGCCCGTTGATGCGTGTCGCGAATCGGGTCGGGGAGGCGAATCATTGAAAA 1771 CI\GTAACGCCATAAACI\GCGCTTCATTGAGCGGTGCGGCCTGCATCI\CGTGGCGCTGCCAGTGACGACCAATTTCCTGGGCGI\GTGGATA 1861 TTCCGCGCACGACTGCGCCCAAATCTGCTGCACAGGGTTGAATTCCGGGGTGATGCGCC acid sequence (Fig. 2.4) suggested 12-TMS topology, a characteristic of the MFS members. Like other MFS permeases, the PnuD protein consisted of six transmembrane a-helical segments, followed by a central hydrophilic cytoplasmic loop, followed by six additional transmembrane a-helical segments. Two E. coli proteins showed high similarity to PnuD: YegT, a putative nucleoside transporter in E. coli K12 (at 46.9 min, 25% identity; spP76417), and CscB, sucrose permease in an E. coli isolate, EC3132 (23% identity; spP30000) (3, 51). No amino acid sequence similarity was seen between the PnuD protein and the known NMN transporter, PnuC (19, 66, 67, 69), even though PnuD and PnuC proteins both have multiple TMS segments (Fig. 2.4). The PnuD protein belongs to the nucleoside:H+ symporter family. The PnuD protein was previously inferred to be a transporter of phenyl propionate permease (PPP) (52); we will argue that this conclusion is incorrect. A multiple-sequence alignment and a phylogenetic tree (Fig. 2.5) revealed that PnuD is sufficiently related to three proteins, YegT, NupG, and XapB, all of which belong to the nucleoside:H+ symporter (NHS) family. Based on this, we suggest that PnuD is normally a nucleoside:!!* symporter. Several shared motifs were identified in the NHS amino acid sequences. Of these, it is noteworthy that a motif within TMS5 was shared by both the NHS and the oligosaccharide :!!* symporter (OHS) proteins. This proposed 12-residue motif, designated as motif 5, is the sequence [(D/E)(T/Y)(P/G)xxRxxGxxG], where x represents any amino acid. The first residue was aspartic acid in the NHS proteins and glutamic acid in the OHS proteins with no exceptions. It is presumed that this slight structural difference may account for their recognition of different classes of substrates, nucleosides 31 acid sequence (Fig. 2.4) suggested 12-TMS topology, a characteristic of the MFS members. Like other MFS perrneases, the PnuD protein consisted of six transmembrane a-helical segments, followed by a central hydrophilic cytoplasmic loop, followed by six additional transmembrane a-helical segments. Two E. coli proteins showed high similarity to PnuD: Y egT, a putative nucleoside transporter in E. coli K12 (at 46.9 min, 25% identity; spP76417), and CscB, sucrose permease in an E. coli isolate, EC3132 (23% identity; spP30000) (3, 51). No amino acid sequence similarity was seen between the PnuD protein and the known NMN transporter, PnuC (19, 66, 67, 69), even though PnuD and PnuC proteins both have multiple TMS segments (Fig. 2.4). The PnuD protein belongs to the nucleoside:H+ symporter family. The PnuD protein was previously inferred to be a transporter of phenyl propionate permease (PPP) (52); we will argue that this conclusion is incorrect. A multiple-sequence alignment and a phylogenetic tree (Fig. 2.5) revealed that PnuD is sufficiently related to three proteins, YegT, NupG, and XapB, all of which belong to the nucleoside:H+ symporter (NBS) family. Based on this, we suggest that PnuD is normally a nucleoside:W symporter. Several shared motifs were identified in the NBS amino acid sequences. Of these, it is noteworthy that a motif within TMS5 was shared by both the NBS and the oligosaccharide:W symporter (OBS) proteins. This proposed 12-residue motif, designated as motif 5, is the sequence [(DIE)(FIY)(P/G)xxRxxGxxG], where x represents any amino acid. The first residue was aspartic acid in the NBS proteins and glutamic acid in the OBS proteins with no exceptions. It is presumed that this slight structural difference may account for their recognition of different classes of substrates, nucleosides P n u D 3- hydrophobic • 0 - hydrophilic - | i i i i I i i i i I i i i i I i i i i P n u C 1 i i i i I i i i i I i i MFS. 32 PnuD 3 hydrophobic 0 hydrophilic - 3 I I I I I I I I I I I I 0 100 200 300 379 PnuC 3 hydrophobic 0 hydrophilic -3 I I I 0 100 200 239 FIG. 2.4. Hydrophobicity plots for the PnuD and PnuC proteins of Salmonella serovar Typhimurium. These were prepared with the PepPlot program (14). The PnuD protein shows extreme hydrophobicity and 12-TMS topology, indicating that it is an integral membrane protein belonging to the :MFS. The PnuC protein also has multiple TMSs, but no amino acid sequence similarity is observed between the two proteins. Since several errors in the database entry M85180 for the pnuC sequence were discovered, the correct­ed amino acid sequence of PnuC was used in this work. 33 E. (spQ47142; TC #2.A.1.27.1), YegT (spP76417), NupG (spP09452; TC # 2.A.1.10.1), and XapB (spP45562; TC #2.A. 1.10.2). The Salmonella serovar Typhimurium genome also carries all four genes for these proteins. (A) A multiple-sequence alignment for the NHS proteins. This was prepared with CLUSTAL W (59). Sequence numbers on the left refer to the position of the leftmost residue on each line. The predicted positions of the TMS shown above the alignment were determined by analysis of protein hydropathy profiles with the PepPlot program (14) and referring to a topological model for XapB (41). Residues conserved in all four and three proteins at any position are in black and gray boxes, respectively. The NHS-specific residues, which are conserved exclusively to the NHS proteins but not found in other MFS proteins including the OHS proteins, are designated by asterisks below the alignment. (B) A phylogenetic tree displaying the relationships among proteins within the NHS family. Phylogenetic analysis was performed with PILEUP (14). Branch points indicate the relative levels of similarity, which increases from left to right. FIG. 2.5. The newly expanded NHS family within the MFS. E. coli proteins were used for this analysis; relevant accession numbers and TC numbers of these proteins are PnuD 1. 10.2). XapB ( A ) i TMS1 > < TMS2 > < TMS3 > PnuD 1 MVLQSTRWJ^GY^YFFSYgl-FLPFWSVWLKGI-^ YegT 1 - -||KTTAK jSFMM ^VEWFIW gA||FVPLW-LWLSKS -WSyACTA^WILS^IIiVGSI-Tj -RFFS^KVLAVXMFAGgLLMYFAgQQTTFA NupG 1 --JMNLKLQ JKILS iLQFCLWgS^TTi|LG-SYMFWLKI*DGAS§^V^SSL^A^FMPA^DSIV--Ayj --KWLsBKWVYAICHTIGglTLFMBSQVTTPE XapB 1 - -^SIAMRgKVMSaLQYf I||gS^VT|G-SYMINTLI^TGANV^v|sSK||||gI IM|GIM|l I - -Ag- -KWI^^RAYMLCHLVCgGVLFYIgSpPDPD * ** * * * * * * * ** * * * * ** < TMS4 » < TMS5 > < TMS6 > PnuD 95 WLMLyMIGFNLFFSgLVPgTDALANTWQKQF PLDYGKVgLWfflsVAFVIGSALTGKLVTMFDYRVI LAJjfcTLGVA^lflGFLIRPTIQPQGAS YegT 93 GF|pLLLAYSLTYM JjCIAIrjSIAFANVPDV ERDFPRIJVMgTIGWIASGLACGFLPQILGYADISPTNIPjjLSTAGS' ALJG-VFAFFLPDTP-P NupG 94 AMFLVILINSFAYW VTLG JltJTISYYRLQNAGMDIVTpFPPI JlWgTIGFIMAMWWS LSGFELSHMQ JYIGAAL' AI Jv-LFTLTLPHIPVA XapB 94 M^^^VKAMAF^^IAQSgSVSYSCLAQAGLDPVTA^P|^VFg§V^^|VAMWAVS LLHI^LSSL^iASGA^Ss-AYALfSlKIgVA * * * * * * * * ** * * * ** * < TMS7 > < TMS8 > < PnuD 187 RQQEST ^S^WLALVRQNWR-RLACVCLLQGAHAAYYGRSAIYWQAAG YSASAVGYLWSLGVVAQVIIFALSNKLFR^CSARDMIIL YegT 187 KSTGKMDIKVML gLDgLILLRDKNFLV aFFCSFLFAMPLAFYYI gANGYLTEVG MKNATGWMTLGQFS a IFFMLALPFFTKgFGIKKVLL NupG 186 KQQANQSWTTIX gLDgFA^ XapB 186 EKKATTSIASKLgL^FVLFKNPRMAI^LFA **** * * * * * * * *** ** _ TMS9 > < TMS10 > < TMS11 > PnuD 272 ISAICGWgWGIMGATTALPW ffllWQILHCGTFWCHIAAMRYIAARQGSEV-IRlfflAVYSAVA^GSIAIMWFAHFLYQ YegT 277 LGLVTAAIJYGFFIYGSADEYFTYALJF^GILLHCVSYDFYYVTAYIYYDKKAPVHMRTAft gGLITLCCQ e FGSLLGYRLG e VMMEKMFAYQEPVNGLTF NupG 286 I § I ^AWI L yFALFAYGD PT PFGT - Vt gVpSMIVYGCAFDFFN IS GSVFVEKEVS PAI RASA JX3MFLMMTN e F^CILGGIVS eKWE-MY TQNGIT-XapB 286 MgMiAWTLEFgFg^^^ SVDGVK- * * * ** * * * * * * * * * * * * i TMS12 > PnuD 354 -YLGHGVFWVMALVA|,PAMgL RPKWPSC YegT 377 NiSGt^FG|VMIAII§VL^IF|RESDNEITAl|VDDRDIALTQGEVK NupG 379 D|QTV|LIFiGYSVV^FA*4AMpYKHV RVPTGTQTVSH - XapB 379 DgQTIgLVFgGYALF§VIgFFGPKYNHD PE||lKHRAVTH (A) YegT NupG XapB PnuD YegT NupG XapB PnuD YegT NupG XapB PnuD YegT NupG XapB PnuD YegT NupG XapB ( TMSI ) ~( ----TMS2) ( TMS3 ) 1 MVL. QS~RiY FF.sII.-•...•. FLP.FW... SVWLKGI -GL...... T PET~.. LLGA-GL~FLGSLLIAPRV~' PSRLISr"LRV1JU,LTLL~AFWA~HVA--- 1 --HKTT S WiI ",FVP~W-LWLSKS-GFSAGEI S~:ACT..' A..d.. "IL SP.. ILVG.... S I-- '--RFF~.•...~...•.V LMFA~LLM.Y. ,~ QT.TFA 1 --~LK 1LS QFCL S~TT~G-SYMFVT~GASI V1SSL~+ ' FM~~LG1V-- '--KWL~AICHT1~1TL 'VTTPE 1 --MS1 S QYil1 S)lLVTLG-SYMINTLHlT :SSK '1nqG1MiII-- '--~RAYMLCHLV~VLFY 'S~PD * * * * * * * * * * ** * * * * ** ( TMS4 ) ( TMSS) (TMS6) 93 GFJ!P~LJ.tAYSLT ",I :,S1AFANVPDV----ERf>FPRI" . I~~SGLACGFLPQ1LGYAD1SPTN1 .lJJ.: ;TAGS -VFAFFLPDTP-P 95 WLMLYM1GFNLFFS" ViALANTWQKQF----PLQYG "L SVAiV1GSALTGKLVTMFDYRV1----!TLGV FL1RPT1QPQGAS 94 ~¥1~INSFA ",L 1~T1SYYRLQNAGMD1V11l~F.P~" 1,.1, . .J.MAMWVVS-------LSGFELSHM ,1;~u.t:;1R -LFTL'J:'~PH1PVA 94 MMfiHMVN " I HSVSYSCLAQAGLDPVTAUP" .• AMWAVS-------LLHLELSSL iAs\:I.ftt~!!l,S-AYALr!tiK1WA 187 187 186 186 272 277 286 286 354 377 379 379 * ( TMS7) ( TMS8 ) i-- RQQEST--~..S ~."V RQNWR••• iC VC~.'QGAHAA.Y ISA,.1YW.. Q AAG--------YSASAVGYLW.S.. , LGVViIIFA..L S N...K..L I . SA..R..D ~. KSTGKMD1 'LILLRDKNFL FCSFLFAMPLAFYYI GYLTEVG----------MKNATGWMTLGQFS IFFMLALPFFT -FGIKKVLL KQQANQSWTT. , 'FA!:FKNKRMAI.IFS~GAEL,QIT :,; ¥SFDKDPMFASSFIVQHASIIMSISQIS LFI~TIPF~LS" GIKNYMM EKKATTSLAS, , 'FViiFKNPRMA1 F~VjQIT P¥DFARNPEFADSFVVKYPS1LL§VS, GllliflTIIB "FiRTH * ** * * * * * * * - TMS9 ) ( TMSI0) ( TMSll ) I§.A.. I .C§I!liIM...G .A.. T .. TALpw. mj LGLVTAAI" GF('IYGSADEYFTY FIVL... GVIQ L....L..I..... LHHG.CV. GSTY.FD..•. T..F..V.Y.. C YHVTLAAYA1YMVRDYKKI.AAPA\TRHQMGRTS EV••. -...1. VLi 1YTSLACVC FGS.. SILALIGMYRTLV FI§LYKQMF-A--Y-Q-E--P-V-N--G.. -L-T-F I§IYAWI "FAr.~Al'§oPTPFGT- :.)MIVYQcAfP~FNISGSV~KEVSPAlJiAS .• , FLMMT F§CIL';?GIVS -MY----TQNGIT- ~WT "qiFIAIiDPSTTGF-1: : sMjVY~FNISGsvE1EQEVDSSlIAS .L iAWVfjSILS VD-YF----SVDWK- ** * ** ( TMS12 ) ;j~~~~~·~'·'·~·fi-;~;~;;~;~~~:~~;~;~ DIQTVlL1ilGYSVVIIF YKHV---RVPTGTQTVSH------ DIQTIIL~IGYALFiiVI YNHD---PEIIKHRAVTH------ * * * * * P n u D Y e g T N u p G X a p B 35 (=B~) _________ PnuD ~-------------------YegT r--------- NupG L ______ XapB oligosaccharides, Argl33 143 NMN sequence' a aal33 mM mM pnuD+ CGT (Arg) Arg) C^dEE) Arg) pnuD*135 gjGT(S0) CGG (Arg) TT15565 pnuD*136 QGT ( g f ) CGG (Arg) TT15586 pnuD*136 pnuD145 QGT (QJB) Q G G (QBE) 36 and oligo saccharides, respectively. The Arg133 in PnuD (described below) and Gly residues in the motif were highly conserved in two family proteins with alanine being the only observed substitution for glycine, the last residue in the motif. Conservation of the motif 5 within two close families and its absence from other MFS proteins suggest that it plays an important role in interacting with their substrates. Taken together, the observations allowed the PnuD protein to be placed in the NHS family. Sequence changes caused by pnuD* mutations. The all three sequenced pnuD* mutants had missense mutations that altered Arg52, indicating that this Arg residue blocks the ability of the normal PnuD+ transporter to handle NMN (or derived NmR) (Table 2.3). The two pnuD* mutations showing stronger phenotypes changed Arg52 to Ser and Cys, respectively. The weaker mutation pnuD* 143 allowed NMN use with a minimum concentration three times more than the other two did; this pnuD* 143 mutation changed Arg52 to His. The Arg52 residue was the only charged residue in the apparent transmembrane-spanning region TMS2 and thus might be important to substrate TABLE 2.3. Molecular and genetic comparison in the pnuD+ and its mutations Nucleotide and amino acid Growth on NMN Strain Relevant genotype sequencea aa52 aa133 0.1 mM 0.3 mM TT14890 pnuD+ CGT(Arg) CGG (Arg) TT15572 pnuD*143 C~T([[I) CGG (Arg) + TT15564 pnuD*135 ~GT~ CGG (Arg) + + TT15565 pnuD*136 DGT~ CGG (Arg) + + TT15586 pnuD*136 pnuD145 DGT~ [iGGa aThe mutated codons and the respective amino acid substitutions are indicated in black boxes. recognition. Mutations inactivating the PnuD* route of NMN assimilation. To look for additional components of the PnuD* transporter, we started with a strain dependent on PnuD* for NMN assimilation and looked for mutants unable to use NMN. A strain carrying a pncA insertion, a nadA-pnuC deletion, and a pnuD* mutation (TT15565) was mutagenized with DES and survivors were screened for defects in NMN assimilation. All of the NMN" mutations recovered were linked either to a glyA mutation f the pnuD locus) or to a serB mutation (the nadl locus). This is consistent with demonstration that assimilation of NMN via the PnuD* transporter, as for the assimilation via the PnuC transporter, might depend on Nadl protein (see below). No other genes were found that might be required for NMN assimilation. Mutations in the pnuD locus that blocked PnuD*-mediated assimilation of N MN had no phenotype in strains with a functional pnuC gene. Furthermore, intense localized mutagenesis of the pnuD+ gene of a p n u C + strain (TT418) revealed no mutants defective in NMN assimilation. These results support the notion that the PnuD* and PnuC routes are independent modes of NMN assimilation, even though both rely on Nadl(T) function. A pnuD point mutation (pnuD145 in TT15586) that blocked NMN assimilation was isolated in a parental pnuD* mutant (TT15565). It was not genetically separable from the parent pnuD* mutation (in crosses with pnuD+ ) , suggesting that the two mutations are very tightly linked. The mutation changed Argl33 to Trp in the pnuD coding sequence (Table 2.3). The Argl33 residue was central to motif 5 [(D/E)(F/Y)(P/G)xxRxxGxxG] and conserved in all nucleoside (NHS) and oligosaccharide 37 recognition. Mutations inactivating the PnuD* route of NMN assimilation. To look for additional components of the PnuD* transporter, we started with a strain dependent on PnuD* for NMN assimilation and looked for mutants unable to use NMN. A strain carrying a pncA insertion, a nadA-pnuC deletion, and a pnuD* mutation (TT15565) was mutagenized with DES and survivors were screened for defects in NMN assimilation. All of the NMN- mutations recovered were linked either to aglyA mutation (the pnuD locus) or to a serB mutation (the nadI locus). This is consistent with demonstration that assimilation of NMN via the PnuD* transporter, as for the assimilation via the PnuC transporter, might depend on NadI protein (see below). No other genes were found that might be required for NMN assimilation. Mutations in the pnuD locus that blocked PnuD*-mediated assimilation of NMN had no phenotype in strains with a functional pnuC gene. Furthermore, intense localized mutagenesis of the pnuD+ gene of a pnuC+ strain (TT418) revealed no mutants defective in NMN assimilation. These results support the notion that the PnuD* and PnuC routes are independent modes ofNMN assimilation, even though both rely on NadI(T) function. A pnuD point mutation (pnuD145 in TT15586) that blocked NMN assimilation was isolated in a parental pnuD* mutant (TT15565). It was not genetically separable from the parent pnuD* mutation (in crosses with pnuD+), suggesting that the two mutations are very tightly linked. The mutation changed Arg133 to Trp in the pnuD coding sequence (Table 2.3). The Arg133 residue was central to motif 5 [(DIE)(F/Y)(P/G)xxRxxGxxG] and conserved in all nucleoside (NHS) and oligosaccharide (OHS) transporters in the MFS (Fig. 2.5). Thus this conserved residue is critical to NMN (or NmR) transport by PnuD* protein. Dependence of the PnuD* route on NadI(T+ ) function. The Nadl protein was shown previously to have two separable functions (An additional function will be discussed later). The normal transporter of NMN, PnuC, is repressed by Nadl(R) function and requires participation of Nadl(T) function for its activity (66, 67, 69). This was previously interpreted as evidence that Nadl(T) regulates the activity of PnuC protein in response to internal NAD levels. To further test these Nadl effects on PnuD*, a series of nadl alleles were introduced into a pnuD* strain (TT15565) by transduction. This recipient strain also carries a pncA insertion and a nadA-pnuC deletion, and thus relies on its pnuD* allele for growth on NMN. Mutations that eliminate NadI(T+ ) function - (R+T~) and (R~T~), TT15621 and TT15623, respectively - blocked NMN assimilation (Table 2.4). A nadl repressor mutation, (R"T+), in TT15622 did not affect growth on NMN. The nadl NMN ACnadA - TT15620 ACnadA ACnadA nadI(R+T") - ACnadA nadICR"T+) ACnadA nadICR"T") - ACnadA nadICRsT~) + 38 (OHS) transporters in the MFS (Fig. 2.5). Thus this conserved residue is critical to NMN (or NmR) transport by PnuD* protein. Dependence of the PnuD* route on NadI(T+) function. The Nadl protein was shown previously to have two separable functions (An additional function will be discussed later). The normal transporter of NMN, PnuC, is repressed by Nadl(R) function and requires participation of Nadl(T) function for its activity (66, 67, 69). This was previously interpreted as evidence that Nadl(T) regulates the activity of PnuC protein in response to internal NAD levels. To further test these Nadl effects on PnuD*, a series of nadl alleles were introduced into a pnuD* strain (TT15565) by transduction. This recipient strain also carries a pncA insertion and a nadA-pnuC deletion, and thus relies on its pnuD* allele for growth on NMN. Mutations that eliminate Nadl(T+) function - (R+T-) and (R-T-), TT15621 and TT15623, respectively - blocked NMN assimilation (Table 2.4). A nadl repressor mutation, (R-T+), in TT15622 did not affect growth on NMN. The nadl TABLE 2.4. Effect of nadl mutations on the pnuD* phenotype Strain Relevant genotype Growth on 0.1 mM NMN TT14890 pncA ~(nadA pnuC) TTl5620 pncA ~(nadA pnuC) pnuD* + TT15621 pncA ~(nadA pnuC) pnuD* nadl(R+T-) TT15622 pncA ~(nadA pnuC) pnuD* nadl(R-T+) +- TT15623 pncA ~(nadA pnuC) pnuD* nadl(R-T-) TT15624 pncA ~(nadA pnuC) pnuD* nadl(RST-) 39 super-repressor mutation, (RST~), in TT15624 reduced the ability to grow on NMN, but its effect was less severe than observed for the PnuC-mediated transport; this mutant is also defective in the Nadl(T) function (66). These results suggest that the transport function of Nadl is needed for activity of the PnuD*-mediated assimilation. It is suggested here that PnuD normally transports some unrelated compounds and has been recruited for NMN transport by the pnuD* mutation. If this is the case, it seems unlikely that the role of Nadl(T) is an intimate one in controlling PnuD activation. Thus the Nadl requirement seems more to reflect an internal role in NMN assimilation that is critical to both PnuC and PnuD* proteins. A possible role for NadI(T+ ) protein in transport activity of the PnuC and PnuD* transporters is diagrammed in Fig. 2.6. It suggests that the Nadl(T) function could act internally to stimulate assimilation of NMN, regardless of how NmR entered the cell. Expression of the pnuD gene is not influenced by the pyridine concentration or the NadI(R+ ) protein. If PnuD normally transports a pyridine compound, one might expect pnuD gene expression, like pnuC expression, to be regulated in response to pyridine levels. Transcription of the nadB gene and the nadA-pnuC operon is controlled by Nadl(R) function in response to endogenous NAD levels (11, 18, 47, 66, 68, 69). These genes are repressed by Nadl in auxotrophic cells growing with high exogenous nicotinic acid (Na) level (>0.2 mM). Mutations in the repressor domain of Nadl, nadI(R"), cause high expression of nadB and the nadA-pnuC operon even during growth on the Na levels that normally repress. 39 super-repressor mutation, (RsT-), in TTl 5624 reduced the ability to grow on NMN. but its effect was less severe than observed for the Pnue-mediated transport; this mutant is also defective in the NadJ(T) function (66). These results suggest that the transport function ofNadl is needed for activity of the PnuO*-mediated assimilation. It is suggested here that PnuD normally transports some unrelated compounds and has been recruited for NMN transport by the pnuD· mutation. If this is the case, it seems unlikely that the role of NadJ(T) is an intimate one in controlling PnuD activation. Thus the Nadl requirement seems more to reflect an internal role in NMN assimilation that is critical to both Pnue and PnuD* proteins. A possible role for Nadl(T) protein in transport activity of the Pnue and PnuD* transporters is diagrammed. in Fig. 2.6. It suggests that the NadI(T) function could act internally to stimulate assimilation of NMN. regard less of how NmR entered the cell. Expression of the pnuD gene is not influenced by the pyridine concentration or the Nadl(R·) protein. If PnuD normally transports a pyridine compound, one might expect lllli!I2gene expression, like.I!lli!C expression, to be regulated in response to pyridine levels. Transcription of the nadS gene and the nadA.ll.!ll:!C operon is controlled by Nadl(R) function in response to endogenous NAD levels (II, 18, 47, 66, 68, 69). These genes are repressed by Nadl in auxotrophic cells growing with high exogenous nicotinic acid (Na) level (>0.2 mM). Mutations in the repressor domain of NadJ, nadl(R-). cause high expression of nadS and the nadA-lID..J,!C operon even during growth on the Na levels that normally repress. G e n o t y p e P h e n o t y p e M o d e l P n u C P n u D Nadl + + G r o w t h 0.1 m M N MN T+ ) NR -NMN cytoplasm + periplasm N R PnuC"*" J"jR' QNadI(T") V ^ PnuD NR NR + + + NR' > PnuC o NR7 NR NR " NR'" NR' NR NMN NR ^'nuD* NR NMN NR transported by either PnuC or PnuD*. The Nadl protein might play a catalytic role in converting transported NR into NMN in cytoplasm. Like the PnuC route, the PnuD* route required NadI(T+) function for NMN assimilation. 40 Genotype Phenotype Model PnuC PnuD NadI Growth on 0.1 mMNMN NadI(T+) NR 8 .. NMN + + + + cytoplasm • periplasm NR PnuC+ NR~nuD+ ~I(T) NR NMN • NR'- + + NR • + + 0 + NR * + + NR~ NR e. NMN ~uD' NR~NMN * NR~ N~R FIG. 2.6. Dependence of the PnuC and PnuD* proteins on NadI(T+) function. It is assumed that exogenous NMN is degraded first in periplasm and the produced NR is NadI playa N adI(T +) To pursue the function of PnuD+ , MudJ insertions in pnuD (see above) were used to test regulation of the pnuD gene. A strain with a nadB deletion and a pnuD::lac fusion (TT20241) was grown on various concentrations of Na and NMN and with various nadl mutations. As a control, a strain with a pnuC::lac fusion (TT14938) was also tested. Variations in the level of exogenous Na and NMN and several nadl mutations (TT15621 to TT 15624) had the expected effects on the control pnuC::lac fusion, but did not alter pnuD gene transcription level. Previous work suggested that a consensus sequence referred to as the NAD box consisting of TGTTTA and its inverted repeat separated by 5 to 6 bp is the Nadl repressor binding site (46, 47). The known Nadl(R) target genes (nadB, nadA-pnuC and pncB) from both Salmonella serovar Typhimurium and E. coli all share this NAD box near their promoters. Consistent with the results above, the NAD box was not found in the pnuD control region (Fig. 2.3). Evidence for a role of PnuD+ in pyridine transport was also sought by testing resistance to the pyridine analogs, 6-aminonicotinamide and 6-aminonicotinic acid. (25). These analogues inhibit growth after being converted by PncA, PncB, NadD, and NadE to the level of NAD. It seems possible that pnuD null mutations might provide a measure of resistance if PnuD contributes pyridine transport. No detectable differences were observed in the level of resistance among the wild-type (TR10000), a pnuD strain (TT20240), a pnuC strain (TT22862), and a p n u D pnuC strain (TT22865). These results suggest that PnuD+ does not normally transport pyridines such as Na and Nm. Native substrate(s) of the P n u D + t r a n s p o r t e r . Null mutations in the pnuD gene were tested for their effect on assimilation of various pyridine compounds. This was 41 To pursue the function ofPnuD+, Mud] insertions in pnuD (see above) were used to test regulation of the pnuD gene. A strain with a nadB deletion and a pnuD::lac fusion (TT20241) was grown on various concentrations ofNa and NMN and with various nadl mutations. As a control, a strain with a pnuC::lac fusion (TT14938) was also tested. Variations in the level of exogenous Na and NMN and several nadl mutations (TT15621 to TT15624) had the expected effects on the control pnuC::lac fusion, but did not alter pnuD gene transcription level. Previous work suggested that a consensus sequence referred to as the NAD box consisting of TGTTTA and its inverted repeat separated by 5 to 6 bp is the Nadl repressor binding site (46, 47). The known Nadl(R) target genes (nadB, nadA-pnuC, and pncB) from both Salmonella serovar Typhimurium and E. coli all share this NAD box near their promoters. Consistent with the results above, the NAD box was not found in the pnuD control region (Fig. 2.3). Evidence for a role of PnuD+ in pyridine transport was also sought by testing resistance to the pyridine analogs, 6-aminonicotinamide and 6-aminonicotinic acid. (25). These analogues inhibit growth after being converted by PncA, PncB, NadD, and NadE to the level ofNAD. It seems possible that pnuD null mutations might provide a measure of resistance if PnuD contributes pyridine transport. No detectable differences were observed in the level of resistance among the wild-type (TR10000), a pnuD strain (TT20240), a pnuC strain (TT22862), and a pnuD pnuC strain (TT22865). These results suggest that PnuD+ does not normally transport pyridines such as Na and Nm. Native substrate(s) of the PnuD+ transporter. Null mutations in the pnuD gene were tested for their effect on assimilation of various pyridine compounds. This was done to look for evidence that the PnuD+ protein might transport some pyridine compound and be adapted to NmR transport by the pnuD* mutations. As seen in Table 2.5, however, a pnuD null mutation did not impair growth on pyridine compounds tested (Nm, NMN, Na, and NaMN). Seeking mutations that make the PnuD* t r a n s p o r t e r independent of NadI(T+ ) . Previous work on the pnuC gene showed that its dependence on Nadl(T) function could be eliminated by point mutations within the pnuC coding sequence (66). More recently it has been shown that PnuC actively transports NmR and that mutant forms able to transport intact NMN become independent of Nadl(T) function (U. Bergthorsson, Y. Xu, J. Sterneckert, B. Khodaverdian, H. Jeong, and J. R. Roth, unpublished data). This suggests that Nadl(T) is a dispensible feature of NmR transport mechanism and is only needed to assure retention or assimilation of NmR following import. To test whether PnuD* can also become independent of Nadl(T) function, a pnuD* mutation was placed with a nadl(T) mutation and the double mutant was placed under selection for derivatives that restored ability to grow on NMN. A strain carrying a p n cA insertion, a nadA-pnuC deletion, a pnuD* mutation, and a nadl insertion (TT23093) was mutagenized with N-methyl-N'-nitro-N-nitrosoguanidine and revertants restoring ability to use 0.1 mM NMN were selected. All six of the observed revertants, however, did not retain the pncA insertion mutation and were phenotypically PncA+ ; the revertants were able to grow on 10 uM NMN. No other types of revertants were found, suggesting that it is not simple to convert PnuD* into a transporter of intact NMN. 42 done to look for evidence that the PnuD+ protein might transport some pyridine compound and be adapted to NmR transport by the pnuD* mutations. As seen in Table 2.5, however, a pnuD null mutation did not impair growth on pyridine compounds tested (Nm, NMN, Na, and NaMN). Seeking mutations that make the PnuD* transporter independent of NadI(T+). Previous work on the pnuC gene showed that its dependence on NadI(T) function could be eliminated by point mutations within the pnuC coding sequence (66). More recently it has been shown that PnuC actively transports NmR and that mutant forms able to transport intact NMN become independent of NadI(T) function (U. Bergthorsson, Y. Xu, J. Stemeckert, B. Khodaverdian, H. Jeong, and J. R. Roth, unpublished data). This suggests that NadI(T) is a dispensible feature of NmR transport mechanism and is only needed to assure retention or assimilation of NmR following import. To test whether PnuD* can also become independent of NadI(T) function, a pnuD* mutation was placed with a nadI(T) mutation and the double mutant was placed under selection for derivatives that restored ability to grow on NMN. A strain carrying a pncA insertion, a nadA-pnuC deletion, a pnuD* mutation, and a nadI insertion (TT23093) was mutagenized with N-methyl-N'-nitro-N-nitrosoguanidine and revertants restoring ability to use 0.1 mM NMN were selected. All six of the observed revertants, however, did not retain the pncA insertion mutation and were phenotypically PncA\ the revertants were able to grow on 10 ~M NMN. No other types of revertants were found, suggesting that it is not simple to convert PnuD* into a transporter of intact NMN. NaMN 1 uM 0.1 mM 10 uM 0.1 mM 1 mM 1 uM 0.1 mM 10 uM 0.1 mM - - pnuC" + - pnuC" pnuD" - TT 12990 - - - - pnuC+ pnuD" - - - - TT 14890 pnuCT - - - - - - pnuCT - - - - pnuC" pnuD" - - - - - - pnuCT - - - - - - - - pnuC" pnuD" - - - - - - - - TABLE 2.5. Phenotypes of pnuC and pnuD strains on different pyridines Nm NMN Na NaMN Strain Relevant genotype I~M ~M ImM I~M ~M TT9923 nadB pnuC+ pnuD+ + + + + + + + + TT20241 nadB pnuC+ pnuD- + + + + + + + + TT15148 nadA pnuC- pnuD+ + + + + + + + + TT22244 nadA pnuC- pnuD- + + + + + + + + TTl2990 pncA nadB pnuC+ pnuD+ + + + + + TT22245 pncA nadB pnuCt pnuD- + + + + + TT14890 pncA nadA pnuC- pnuD+ + + + TT15565 pncA nadA pnuC- pnuD* + + + + + TT20190 pncA nadA pnuC- pnuD- + + + TT22246 pncB nadA pnuC- pnuD+ + TT22247 pncB nadA pnuC- pnuD- + 44 Differences between Salmonella serovar Typhimurium and E. coli in the pnuD region. The S. enterica pnuD gene is similar to the hcaT gene of E. coli; the amino acid sequences were 88% identical. The genes map at analogous chromosomal positions. However, the Salmonella pnuD gene is 195 bp from glyA. whereas in E. coli a cluster of genes (16 kb) lies between the pnuD and glyA genes. Two regions are compared in Fig. 2.7. About 5.4 kb of this intervening region (the hca. cluster) was characterized by Diaz et al. (15) and shown to encode the dioxygenolytic pathway for catabolism of 3- phenylpropionic (hydrocinnamic) acid. The region includes an operon of five catabolic genes fhcaA!A2CBD) and a divergently transcribed regulatory gene (hcaR). The nearby pnuD orthologue was named hcaT in the expectation that this gene product might be a transporter of 3-phenylpropionic acid (3-PP); no data supporting this transport role or importance of HcaT to use of 3-PP were provided. Based only on its proximity to the hca genes, HcaT (PnuD) has been assigned a function in phenyl propionate transport. Since Salmonella serovar Typhimurium lacks the hca operon and cannot utilize 3- phenylpropionic acid (15), it is unlikely that the retained pnuD gene is a part of the hca system in E. coli or Salmonella. Genetic mapping of the pnuP region. In addition to the pnuD mutations described above, a second type of mutation (pnuP) restored ability to use NMN to strains, in which known assimilation routes were blocked. To locate the pnuP gene on the genetic map, a TniOdTc insertion (zxx-9180::Tn 1 OdTc in TT23104), 76% cotransducible with pnuP*144. was isolated as described in Materials and Methods. The sequence flanking of this insertion was determined as described in Materials and Methods. Differences between Salmonella serovar Typhimurium and .Eo coli in the pnuD region. The S. enterica pnuD gene is similar to the hcaT gene of :E. coli; the amino acid sequences were 88% identical. The genes map at analogous chromosomal positions. However, the Salmonella pnuD gene is 195 bp from glyA, whereas in :E. coli a cluster of genes (16 kb) lies between the pnuD and glyA genes. Two regions are compared in Fig. 2.7. About 5.4 kb of this intervening region (the hca cluster) was characterized by Diaz et al. (15) and shown to encode the dioxygenolytic pathway for catabolism of 3- phenylpropionic (hydrocinnamic) acid. The region includes an operon of five catabolic genes (hcaAIA2CBD) and a divergently transcribed regulatory gene (hcaR). The nearby pnuD orthologue was named hcaT in the expectation that this gene product might be a transporter of 3-phenylpropionic acid (3-PP); no data supporting this transport role or importance of HcaT to use of 3-PP were provided. Based only on its proximity to the hca genes, HcaT (PnuD) has been assigned a function in phenyl propionate transport. Since Salmonella serovar Typhimurium lacks the hca operon and cannot utilize 3- phenylpropionic acid (15), it is unlikely that the retained pnuD gene is a part of the hca system in E.. coli or Salmonella. Genetic mapping of the pnuP region. In addition to the pnuD mutations described above, a second type of mutation (pnuP) restored ability to use NMN to strains, in which known assimilation routes were blocked. To locate the pnuP gene on the genetic map, a Tnl.QdTc insertion (zxx-9180::Tnl0dTc in TT231 04), 76% cotransducible with pnuP* 144, was isolated as described in Materials and Methods. The sequence flanking of this insertion was determined as described in Materials and Methods. of the E. coli. The pnuD gene was immediately adjacent to the glyA gene Salmonella serovar Typhimurium, but the E. coji corresponding region included about 16 kb insert between these two genes. (A) Spatial organization comparison. The intervening region of E. coh includes the hca operon of six genes and the following uncharacterized region. Arrows sequence similarity. E. coli. which showed no significant similarity with the pnuD-glyA gap in Salmonella serovar Typhimurium. Nucleotide +1 denotes the initiator of the pnuD gene. The pnuD start codon and the hcaR and glyA stop codons are designated by the asterisks. For each sequence, the most probable transcription start site is indicated as the symbol ( • ), together with a putative promoter region (-35 and - 1 0 ) and a putative ribosome binding site (RBS) as bold styles. Stm, Salmonella serovar Typhimurium; Eco, E. coli. 45 FIG. 2.7. Comparisons afthe pnuD-glyA region in Salmonella serovar Typhimurium and pouD g]yA in Typhimuriurn, hut &. coli }3. coli indicate the transcriptional orientations of the respective genes. (B) Nucleotide se:quence alignment of similarity, The dotted line indicates the 16 kb intervening region in £. coli, + pnuQ heaR g,lyA sequence. symb()l (.), with 10) hinding RES) ( A ) Salmonella serovar Typhimurium E. coli hca ( B ) start -35 RBS • -10 TG TCCGCCTCCCTCTTCAAGTTTGCCGT S tm +15 CCTTACGTTACGGTAtTAAGGGAGGGCAAACCTCCGCTATCAGACCGCCATTTGCCAACGCGACCGTTC-CCATTGTATTCCAAT III Mill 111111 l l l l l l l l l l 1 IIIIMM I Ml II II Mill I II Mill II +15 CCTAACGTTTTGGTACTCAGGGAGGGCAGTCCTCCGCAATCAGACCCCAATTAGCTTACAACCACGTTCAACTTTTTATTCTTAT RBS A AGCTCGGACGAAGG 50 bp Stm -98 TGCGGCGGCCTATTACTGAAGGCCGCCGCAGTGCAATACCGCGTATTCATTACGGGCTGTCTGCGTCGCGGTAGCCCGGTAAAAGTGTTGTGTT Eco -136 GCGTACAACGTTTCTATTACGCATT AATACGGCAATGCGAACGGTTTGCAAGGAC -10 GCG|\ATACGCATTTGGCC I i 1111 M 11111 GCGkATACGCATTTGGCC 16 kb insert TCGTTTAGTGGCAA -35 end hcaR end (A) (B) E. coli operon pnuD start ..... csiE GTCCGCCTCCCTCTTCAAGTTTGCCGT * RBS • -10 T Stm +15 CCTTACGTTACGGT TAAGGGAGGGCAAACCTCCGCTATCAGACCGCCATTTGCCAACGCGACCGTTC-CCATTGTATTCCAAT III 11111 111111 1111111111 1111111 11111111 I III II II 11111 I II 11111 II Eco +15 CCTAACGTTTTGGT TCAGGGAGGGCAGTCCTCCGCAATCAGACCCCAATTAGCTTACAACCACGTTCAACTTTTTATTCTTAT ____________ ~* RBS A AGCTCGGACGAAGG--- --- TGCGGCGGCCTATTACTGAAGGCCGCCGCAGTGCAATACCGCGTATTCATTACGGGCTGTCTGCGTCGCGGTAGCCCGGTAAAAGTGTTGTGTT C * A GCGTACAACGTTTCTATTACGCAT TACGGCAATGCGAACGGTTTGCAAGGAC- --- ------TCGTTTAGTGGCAA • * heaR end GCG TACGCATTTGGCC 111111111111111111 GCG TACGCATTTGGCC * glyA end 47 Database searches revealed that the insertion was located in the ydeE (32.6 min) coding sequence; this insertion was renamed as vdeE137::Tnl0dTc. Transduction tests with a few genetic markers in this region indicated that the pnuP* mutation is in or very close to either ydeA or marC at 32.7 min. The inferred structure of the region is diagrammed in Fig. 2.8. The pnuP*144 mutation is a missense mutation at the motif C of t h e ydeA gene. The sequence of the ydeA gene and the mar operon was determined for the pnuP*144 mutant (TT23524) and for a pnuP+ strain (TT23525) as described in Materials and Methods. Strain TT23524 had a missense mutation that altered Vail51 to Gly in the amino acid sequence (GTG to GGG in the nucleotide sequence) of YdeA. The PnuP (=YdeA) protein is known to export L-arabinose and isopropyl-|3-D-thiogalactopyranoside (IPTG) in E. cpH (4, 5). The closest characterized homologues of PnuP (Fig. 2.9) were i) OpdE of Pseudomonas aeruginosa (24), ii) chloramphenicol efflux pumps, Cmx (58), CmlB (13), and CmlR (16), and iii) AraJ, arabinose efflux pump, of E.