<%BANNER%>

The Functions of Streptococcus mutans YidC1 and YidC2, and their Roles in Membrane Biogenesis

Permanent Link: http://ufdc.ufl.edu/UFE0042762/00001

Material Information

Title: The Functions of Streptococcus mutans YidC1 and YidC2, and their Roles in Membrane Biogenesis
Physical Description: 1 online resource (208 p.)
Language: english
Creator: PALMER,SARA R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: BNPAGE -- MEMBRANES -- SECRETION -- SECY -- SRP -- STREPTOCOCCUS -- TRANSPORT -- YIDC
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Streptococcus mutans is a major causative agent of dental caries. It has two paralogs of the YidC/Oxa1/Alb3 family of membrane protein insertases/chaperones. Like disruption of the signal recognition particle (SRP) co-translational pathway, elimination of yidC2 causes loss of genetic competence, stress sensitivity (acid, osmotic and oxidative), and impaired biofilm formation. It has been postulated that acid sensitivity in these mutants is due to defects in assembly of the F1Fo ATPase, which pumps protons out of the cell during acidic growth conditions. Elimination of yidC1 is less severe, with no observable effect on growth or stress sensitivity. YidC2 has a longer, more positively-charged C-terminal tail than YidC1 and can complement an Oxa1 mutant in yeast. Oxa1 function depends on its C-terminal tail. Chimeric proteins were constructed to evaluate, the roles of the C-termini in stress tolerance of YidC1 and YidC2. Placing the YidC2 tail on YidC1 restored acid and osmotic stress tolerance to a yidC2 mutant. In contrast, placing the C-terminal tail of YidC1 on to YidC2 resulted in a dominant negative effect, with decreased growth yield and sensitivity to stress, indicating the C-terminal domains play an important role in stress tolerance. Studies revealed both YidC1 and YidC2 can function in E. coli to assemble an active F1Fo ATPase in a YidC-depletion strain. In S. mutans, deletion of either yidC1or yidC2 results in decreased membrane-associated ATPase activity, indicating that the pronounced acid sensitive phenotype of the delta yidC2 mutant stems from further mechanisms in addition to impaired proton extrusion. Blue-Native PAGE, which separates native membrane protein complexes, combined with LC-MS/MS revealed differences in location of glycolytic enzymes, citrate metabolic enzymes, transporters, and ribosomal proteins associated with the membrane fractions of wildtype and the yidC mutants. Therefore, YidC1 and YidC2 appear to contribute to assembly of membrane-associated complexes involved with glycolysis and to ribosome tethering for co-translational translocation, likely explaining the dispensability of the SRP pathway in S. mutans. Collectively, these results provide evidence that the difference in stress sensitivity between the delta yidC1 and delta yidC2 mutants stems from functional differences in their C-termini.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by SARA R PALMER.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brady, L. Jeannine.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-10-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042762:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042762/00001

Material Information

Title: The Functions of Streptococcus mutans YidC1 and YidC2, and their Roles in Membrane Biogenesis
Physical Description: 1 online resource (208 p.)
Language: english
Creator: PALMER,SARA R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: BNPAGE -- MEMBRANES -- SECRETION -- SECY -- SRP -- STREPTOCOCCUS -- TRANSPORT -- YIDC
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Streptococcus mutans is a major causative agent of dental caries. It has two paralogs of the YidC/Oxa1/Alb3 family of membrane protein insertases/chaperones. Like disruption of the signal recognition particle (SRP) co-translational pathway, elimination of yidC2 causes loss of genetic competence, stress sensitivity (acid, osmotic and oxidative), and impaired biofilm formation. It has been postulated that acid sensitivity in these mutants is due to defects in assembly of the F1Fo ATPase, which pumps protons out of the cell during acidic growth conditions. Elimination of yidC1 is less severe, with no observable effect on growth or stress sensitivity. YidC2 has a longer, more positively-charged C-terminal tail than YidC1 and can complement an Oxa1 mutant in yeast. Oxa1 function depends on its C-terminal tail. Chimeric proteins were constructed to evaluate, the roles of the C-termini in stress tolerance of YidC1 and YidC2. Placing the YidC2 tail on YidC1 restored acid and osmotic stress tolerance to a yidC2 mutant. In contrast, placing the C-terminal tail of YidC1 on to YidC2 resulted in a dominant negative effect, with decreased growth yield and sensitivity to stress, indicating the C-terminal domains play an important role in stress tolerance. Studies revealed both YidC1 and YidC2 can function in E. coli to assemble an active F1Fo ATPase in a YidC-depletion strain. In S. mutans, deletion of either yidC1or yidC2 results in decreased membrane-associated ATPase activity, indicating that the pronounced acid sensitive phenotype of the delta yidC2 mutant stems from further mechanisms in addition to impaired proton extrusion. Blue-Native PAGE, which separates native membrane protein complexes, combined with LC-MS/MS revealed differences in location of glycolytic enzymes, citrate metabolic enzymes, transporters, and ribosomal proteins associated with the membrane fractions of wildtype and the yidC mutants. Therefore, YidC1 and YidC2 appear to contribute to assembly of membrane-associated complexes involved with glycolysis and to ribosome tethering for co-translational translocation, likely explaining the dispensability of the SRP pathway in S. mutans. Collectively, these results provide evidence that the difference in stress sensitivity between the delta yidC1 and delta yidC2 mutants stems from functional differences in their C-termini.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by SARA R PALMER.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Brady, L. Jeannine.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-10-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042762:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE FUNCTION OF S treptococcus mutans YIDC1 AND YIDC2, AND THEIR ROLES IN MEMBRANE BIOGENESIS By SARA MARIE RASER PALMER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Sara Marie Raser Palmer

PAGE 3

3 To everyone who believed in me and helped along the way, this would not have been possible without you

PAGE 4

4 ACKNOWLEDGMENTS First I would like to thank my h Ellis. He was the first person to recognize my potential as a scientist and without his careful mentoring and passion for teaching I wou ld not have pursued a career in science but would probably be a struggling artist instead. M y A unt and Uncle Carolynn and Terry Palmer, who took me in and gave me a chan ce to make something of my life, have my sincerest appreciation. They taught me the importance of education and helped me to realize my potential as a person I would also li ke to thank my Aunt Anne Palmer, who was there whe I am grateful to Dr. Linda Ma nsfield who was my mentor for undergraduate research Her lab was the first of many labs I hope to work in during my career as a scientist. She gave me a chance to see what science is all about and without her letter of recommendation graduate school wo uld not have been possible I want to acknowledge Dr. Jeannine Brady, my dissertation mentor Jeannine has She has shown compassion, understanding, and an amazing ability to handle all that life throws at you Jeannine has been an excellent role model and has an incredible scientific mind. I am grateful that I chose her as my mentor for so many reasons, but mostly because she challenged me to think for myself and allowed me the freedom to persue my interests I appreciate all the opportunities she has imparted on me thro ugh the year, and I will miss working with her. I was also fortunate to know Dr. Arnold Bleiweis He was a tremendous person and scientist and I feel much honored to have known him and benefited from his vast knowledge and understanding of Streptococcus mutans I am particularly grateful of the time he took to proof read my pre doct oral fellowship

PAGE 5

5 application as well as for all the p apers he printed with me in mind I would like to thank Paula Crowley, for all the advice and knowledge she has imparted over the past five years I particularly want to thank her for help with proof reading my pre doctoral fellowship application as well as all the constructive criticism she has given me in the past I am also indebted to Dr. Lin Zeng for his advice and help with cloning in Streptococcus mutans which was integral to the completion of this project. Last but not least, I want to acknowledge my committee members Thank you for attending my committee meetings and providing lively discussions about my project, for asking tough questions and motivating me to be a better scientist Thank you for taking the time to read this document and for all your criticisms, which have made it better. You have been a great committee and I strive to make you proud In particular, I want to thank Dr. Brian Cain for the use of the equipment in his l ab as well as his advice on the ATP hydrolysis assays He has always been extremely supportive of his students and a great mentor I am also appreciative for the letter of support he wrote on my behalf for my pre doctoral fellowship I owe many t hank s to Dr. Ken Cline for his letter of support and for help with the Blue Native PAGE experiments Dr. Robert Burne for all his advice with my cloning projects and taking the time to meet to discuss my project on several occasions I thank Dr. Paul Gulig for his insistence on excellence and dedication as a teac her in m icrobial genet ics a s well as always attending the immunology and m icrobiology journal clubs. His insight and contributions to these discussions have been invaluable to my education and development as a scientist.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIA TIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 LITERATURE REVIEW ................................ ................................ .......................... 18 Streptococcus mutans and Dental Caries ................................ ............................... 18 Acid Tolerance ................................ ................................ ................................ ........ 21 Secretion and Membrane Protein Biogenesis ................................ ......................... 27 Stress Response and Membrane Biogenesis in S. mutans ................................ .... 32 YidC/Oxa1Alb3 Family of Proteins ................................ ................................ .......... 33 Summary and Specific Aims ................................ ................................ ................... 42 Specific Aim 1: Develop Tools to Examine Compensatory or Redundant Functions in Membrane Biogenesis of S. mutans. ................................ ........ 43 Assembly of the F 1 F o ATP Synthase. ................................ ............................ 43 Specific Aim 3: Determine Differences in Membrane Protein Complexes yidC1 yidC2 Mutants using Blue Native Polyacrylamide Gel Electrophoresis. ................................ ............................. 44 2 DEVELOP TOOLS TO EXAMINE COMPENSATORY OR REDUNDANT FUNCTIONS IN MEMBRANE BIOGENESIS OF Streptococcus mutans ............... 45 Rationale for Study ................................ ................................ ................................ 45 Materials and Methods ................................ ................................ ............................ 47 Synthetic Peptide Antiserum ................................ ................................ ............ 47 Creation of Strains SP13, SP14 and SP17: C Terminal Domain Swaps of YidC1 and YidC2. ................................ ................................ .......................... 50 Conditional Expression of yidC2 ................................ ................................ ....... 52 Deletion Construction of yidC1 using a Spec tinomycin Marker. ..................... 54 Growth Curves ................................ ................................ ................................ 55 Whole Cell Lysates ................................ ................................ ........................... 56 Cell Fractions and Gradient SDS PAGE ................................ .......................... 56 Western Blotting ................................ ................................ ............................... 58

PAGE 7

7 Results and Discussion ................................ ................................ ........................... 58 Terminal Tail C ontributes to Stress Tolerance in S. mutans ........... 58 Conditional Expression of yidC2 ................................ ................................ ....... 62 3 ASSEMBLY OF THE F 1 F O ATP SYNTHASE ................................ ................................ .............. 81 Rationale for Study ................................ ................................ ................................ 81 Materials and Methods ................................ ................................ ............................ 82 Chimeric S. mutans yidC2 yidC E. coli Construction ................................ ........ 82 E. coli YidC Depletion Strain JS7131 with pACYC184 S. mutans yidC1 and yidC2 Constructs. ................................ ................................ .......................... 84 Growth Curves ................................ ................................ ................................ 85 Growth on LB Agar Plates of E. coli JS7131 Expressing S. mutans YidC Proteins. ................................ ................................ ................................ ........ 86 Preparation of Inverted Membrane Vesicles ................................ ..................... 86 ATP Hydrolysis Assays ................................ ................................ .................... 89 Proton Motive Force Assays ................................ ................................ ............. 90 Western Blots ................................ ................................ ................................ ... 91 ATP Hydrolysis Activity of S. mutans Permeablized Whole Cells. .................... 91 Results and Discussion ................................ ................................ ........................... 92 Confirmation of Expression and Membrane Localization of YidC1 and YidC2 Cons tructs in JS7131 ................................ ................................ ......... 92 Restoration of Growth of the E. coli YidC Depletion Strain JS7131 by S. mutans YidC1 and YidC2 in Broth ................................ ................................ 93 Complementation of JS7131 Growth by S. mutans YidC1 and YidC2 on Solid Media ................................ ................................ ................................ ... 94 Rescue of F 1 F o ATPase Activity and PMF in JS7131 by S. mutans YidC1 and YidC2 ................................ ................................ ................................ ..... 95 E. coli YidC can Restore Acid and Salt Tolerance to a yidC2 Mutant in S mutans ................................ ................................ ................................ .......... 97 Involvement of S. mutans YidC1 and YidC2 in Membrane ATPase Activity ..... 98 4 DIFFERENCES IN MEMBRANE PROTEIN COMPLEXES BETWEEN WILDTYPE AND YIDC1 AND YIDC2 MUTANTS USING BN PAGE .................... 119 Rationale for Study ................................ ................................ ............................... 119 Materials and Methods ................................ ................................ .......................... 121 Membrane Fractions and DDM Solublization ................................ ................. 121 Blue Native Polyacrylamide Gel Electrophoresis ................................ ........... 123 Protein Staining of Blue Native Polyacrylamide Gels ................................ ..... 124 In Gel Trypsin Digestion and LC MS/MS (University of Florida, ICBR Core Facility Protocols) ................................ ................................ ........................ 127 BN PAGE with Western Blotting ................................ ................................ ..... 129 Tw o Dimensional BN/SDS PAGE ................................ ................................ .. 130 GAPDH Assays with Whole Cells ................................ ................................ ... 132 Results and Discussion ................................ ................................ ......................... 132

PAGE 8

8 Changes in Protein Complexes in YidC Mutant Membranes were Visible by First Dimension BN PAGE ................................ ................................ .......... 132 YidC1, YidC2 and SecY Co Migrate in High Molecular Weight Complexes ... 133 Differences in Membrane Protein Complexes Between Wild Type NG8 and yidC Mutants in S. mutans Determined by BN PAGE/LC MS/MS ............... 137 Glycolytic enzymes ................................ ................................ .................. 138 Citrate metabolic enzymes. ................................ ................................ ...... 141 Cell wall associated proteins ................................ ................................ .... 142 Transport proteins ................................ ................................ .................... 143 Chaperones ................................ ................................ ............................. 143 Amino acid metabolism ................................ ................................ ............ 144 Butanoate, glutathione, and starch metabolic enzymes ........................... 145 Ribosomal proteins ................................ ................................ .................. 146 Extracellular GAPDH Activity is Increased in yidC Mutants ............................ 148 Discussion ................................ ................................ ................................ ...... 149 5 CONCLUSIONS AND FU TURE DIRECTIONS ................................ .................... 180 Development of Tools to Examine Compensatory or Redundant Functions in Membrane Biogenesis ................................ ................................ ....................... 180 A Function of the C Terminal Tail of YidC2 in Stress Tolerance ........................... 182 YidC1 and YidC2 are Involved in ATPase Assembly ................................ ............ 183 YidC Mutants Sho wed Differences in Membrane Protein Complexes Compare d to Wild T ype NG8 S. mutans ................................ ................................ ............. 183 The Functions of S. mutans YidC1 and YidC2 ................................ ...................... 185 LIST OF REFERENCES ................................ ................................ ............................. 189 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 208

PAGE 9

9 LIST OF TABLES Table page 2 1 Bacterial strains and plasmids used in this study ................................ ............... 70 2 3 Mean doubling times of S. mutans yidC mutants and complemented strains ..... 74 3 1 Complementation of growth of the E. coli YidC depletion strain JS7131 with constructs encoding S. mutans YidC1 and YidC2 ................................ ............ 109 4 1 Summary of proteins identified by LC MS/MS in BN PAGE high molecular weight complexes ................................ ................................ ... 158 _Toc290538357 4 2 Summary of proteins identified by LC MS/MS from BN PAGE lower molecular weight complex ................................ ................................ ................ 160 4 3 Summary of ribosomal proteins identified by LC MS/MS from BN PAGE experiments. ................................ ................................ ................................ ..... 162 4 4 Information regarding proteins identified by LC MS/MS listed in Tables 4 1 to 4 2 ................................ ................................ ................................ .................... 163 4 5 Details of 50S ribosomal proteins indentified by LC MS/MS. ........................... 165 4 6 Details of 30S ribosomal proteins identified by LC MS/MS. ............................. 166 4 7 Glycolytic enzymes identified by LC MS/MS from BN PAGE Band 1 ............... 167 4 8 Glycolytic enzymes identified by LC MS/MS from BN PAGE Band 2 ............... 167 4 9 Glycolytic enzymes identified by LC MS/MS from BN PAGE Band 3 ............... 168 4 10 Glycolytic enzymes identified by LC MS/MS from BN PAGE Band 4 ............... 169 4 11 Proteins identified by LC MS/MS in BN PAGE gel Band 1 (~700 kDa) i nvolved in citrate metabolism ................................ ................................ ........... 170 4 12 Cell wall associated proteins identified by LC MS/MS in BN PAGE ge l Band 1 (~700 kDa) ................................ ................................ ................................ ..... 171 4 13 Cell wall associated proteins identified by LC MS/MS in BN PAGE gel Band 2 (~300 kDa) ................................ ................................ ................................ ..... 171 4 14 Cell wall associated proteins identified by LC MS/MS in B N PAGE gel Band 3 (~70 kDa) ................................ ................................ ................................ ....... 171 4 15 Proteins identified by LC MS/MS in BN PAGE Band 1 (~700 kDa) involved with transport ................................ ................................ ................................ .... 172

PAGE 10

10 4 16 Proteins identified by LC MS/MS in BN PAGE Band 2 (~300 kDa) involved with transport ................................ ................................ ................................ .... 172 4 17 Chaperone proteins identified by LC MS/MS in BN PAGE gel Band 1 ............. 173 4 18 Chaperone proteins identified by LC MS/MS in BN PAGE gel Band 2 ............. 173 4 19 Chaperone prote ins identified by LC MS/MS in BN PAGE gel Band 3 ............. 173 4 20 Chaperone proteins identified by LC MS/MS in BN PAGE gel Band 4 ............. 173 4 21 Proteins identified by LC MS/MS in BN PAGE gel Band 1 (~700 kDa) involved in amino acid metabolism. ................................ ................................ .. 174 4 22 Proteins identified by LC MS/MS in BN PAGE gel Band 2 (~300 kDa) involved with amino acid metabolism ................................ ............................... 174 4 23 Proteins identified by LC MS/MS in BN PAGE gel Band 4 (~64 kDa) involved with amino acid metabolism ................................ ................................ .............. 174 4 24 Proteins identified by LC MS/MS in BN PAGE gel Band 3 (~70 kDa) involved in butanoate, glutathione and starch metabolism. ................................ ............ 175 4 25 Proteins identified by LC MS/MS in BN PAGE gel Band 4 (~64 kDa) involved in butanoate, glutathione and starch metabolism ................................ ............ 175 4 26 Ribosomal proteins identified by LC MS/MS in BN PAGE gel Band 1 ............. 176 4 27 Ribosomal proteins identified by LC MS/MS in BN PAGE gel Band 2 ............ 176 4 28 Ribosomal proteins identified by LC MS/MS in BN PAGE gel Band 3. ............ 177 4 29 Ribosomal proteins identified by LC MS/MS in BN PAGE gel Band 4. ............ 178

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Clustal W sequence alignment of S. mutans YidC1 and YidC2. ........................ 66 2 2 Schematic representations of yidC1 yidC2 and yidC2 of chimeric proteins are shown.. ................................ ................................ ......... 67 2 3 Membrane topology model of YidC1 with location of C terminal peptide used to make C terminal antiserum. ................................ ................................ ............ 68 2 4 Membrane topology model and location of peptides used to make antisera against YidC2 ................................ ................................ ................................ .... 69 2 5 Schematic diagram of the construction of promoter fusions of P celA and P celB with yidC2 ................................ ................................ ................................ 72 2 6 Western blot of whole cell lysates of indicated strains reacted with antibodies against YidC1 an d YidC2 ................................ ................................ .................. 73 2 7 yidC2 yidC2 C S. mutans and complemented mutant strains. ................................ ................................ ............ 75 2 8 Western blots of whole cell lysates from YidC2 depletion strains SP20 and SP10 grown in TDM 0.5% sugar or in THYE.. ................................ .................... 76 2 9 Western blots of while cell lysates from YidC2 depletion strains SP21 and SP11 grown in TDM 0.5% sugar or THYE.. ................................ ........................ 76 2 10 Western blots of membrane fractions from YidC2 depletion strains SP10 and SP20 grown in TDM 0.5% sugar. ................................ ................................ ....... 77 2 11 Cell wall extracts of S. mutans wildtype, yidC mutant strains, or YidC2 de pletion strains SP10 and SP20. ................................ ................................ ...... 78 2 12 Cytoplasmic fractions of S. mutans wildtype, yidC mutant strains or YidC2 depletion strains SP10 and SP20 ................................ ................................ ...... 79 2 13 Membrane fractions of S. mutans wildtype, yidC mutants, or YidC2 depletion strains SP10 and SP20. ................................ ................................ ..................... 80 3 1 Clustal W alignment of the five C terminal transmembrane domains from E. coli YidC and S. mutans YidC1 and YidC2. ................................ ...................... 105 3 2 Predicted membrane topologies of E. coli YidC and S. mutans YidC1 and YidC2. ................................ ................................ ................................ ............... 106

PAGE 12

12 3 3 Confirmation of appropriate production of E. coli YidC and S. mutans YidC1 and YidC2 ................................ ................................ ................................ ......... 107 3 4 Growth curves of E. coli YidC depletion strain JS7131. ................................ ... 108 3 5 Growth on LB agar plates of the E. coli YidC depletion strain JS7131. ............ 110 3 6 ATP hydrolysis activity of E. coli Y idC depletion stra in JS7131. ....................... 111 3 7 Proton motive force (PMF) of E. coli YidC depletion strains JS7131 ................ 112 3 8 Growth curves of S. mutans wild type NG8, yidC1 and with chimeric E. coli 50YidC). ................................ ................................ .................. 113 3 9 Western blot of whole cell lysates of S. mutans wildtype, yidC1 yidC2 and mutant with chimeric E. coli YidC. ................................ ........................ 114 3 10 Membrane associated ATPase activity in S. mutans NG8, yidC1 and yidC2 strains ................................ ................................ ................................ ... 115 3 11 Membrane associated ATP hydrolysis activity of S. mutans wildtype and yidC2 mutant strains ................................ ................................ ....................... 116 3 12 Western blots of membrane samples prepared from S. mutans strains used in ATP hydrolysis assays ................................ ................................ .................. 117 3 13 ATPase activity of S. mutans permeablized whole cells. As say was performed in triplicate ................................ ................................ ...................... 118 4 1 Silver stained versus Coomassie Blue G 250 stained first dimension Blue Native polyacrylamide gels. ................................ ................................ ............. 153 4 2 Silver stained first dimension Blue Native polyacrylamide gel and second dimension Tricine SDS polyacrylamide gels. ................................ .................... 154 4 3 Western blots of Blue Native polyacrylamide gels reacted with antibodies against YidC1, YidC2 and SecY. ................................ ................................ .... 155 4 4 Western blot of second dimension BN/SDS polyacrylamide gels ..................... 156 4 5 Blue Native PAGE showing differences in membrane protein complex composition. ................................ ................................ ................................ ..... 157 4 6 Extracellular GAPDH activity in S. mutans NG8 wild type and various yidC mutants.. ................................ ................................ ................................ ........... 179 5 1 Curren t working model of YidC1 and YidC2 and membrane biogenesis in S. mutans ................................ ......................... 188

PAGE 13

13 LIST OF ABBREVIATION S 2D BN/SDS PAGE 2 nd dimensional Blue Native/ Sodium dodecyl sulfate PAGE AA Amino acids ACMA 9 amino 6 chloro 2 methoxyacridine ACN Acetonitrile ATR Ac id tolerance r esponse BCA Bicinchoninic acid BHI Brain heart infusion BN PAGE Blue Native polyacrylamide gel electrophoresis BSA Bovine serum albumin CAT Chloramphe nicol acetyltransferase DCCD N, N' dicyclohexylcarbodiimide DDM Dodecylmaltoside DTT D ithiothreitol EDTA Ethylenediaminetetraacetic acid ELISA Enzyme linked immunosorbent assay GAPDH Glyceraldehyde 3 phosphate dehydrogenase GTF Glucosyltransferases HK Histidine kinase HPLC High performance liquid chromatography IAA Iodoactamide IPTG D 1 thiogalactopyranoside kDa Kilodaltons LB Luria Bertani LC MS/MS Liquid chromatography MS/MS

PAGE 14

14 LHCP Light harvesting chlorophyll binding protein MS Mass spectom etry MALDI TOF MS Matrix assisted laser desorption/ionization time of flight mass spectrometry MSM Multiple sugar metabolism NICE Nisin inducible controlled expression NT Nucleotide OD Optical density O/N Overnight PAGE Polyacrylamide gel electrophoresis PBS Phosphate buffered saline PCR Polymerase chain reaction PEP PTS Phosphoenolpyruvate phosphotransferase system PFL Pyruvate formate lyase PK Proteinase K PMF Proton motive force Qrt PCR Quantitative real time PCR RBS Ribosome binding site RNAi Ribonucleic Acid Interference RR Response regulator scRNA Small cytoplasmic RNA SGP Streptococcus GTP binding protein SOE PCR Splice overlap extension polymerase chain reaction SRP Signal recognition p article TCA Trichloroacetic acid TDM Tereckyj Defined Media

PAGE 15

15 THYE Todd Hewitt Yeast Extract TIM Translocon of inner membrane TMD Transmembrane domain TOM Translocon of outer membrane WCL Whole cell lysate

PAGE 16

16 Abstract of Dissertation Presented to the Graduate School of the University of Florida i n Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE FUNCTION OF Streptococcus mutans YIDC1 AND YIDC2, AND THEIR ROLES IN MEMBRANE BIOGENESIS By Sara Marie Raser Palmer May 2011 Chair: L. Jeannine Brady Major: Medical Sci ences, Immunology and Microbiology Streptococcus mutans is a major causative agent of dental caries. It has two paralogs of the YidC/Oxa 1 /Alb3 family of membrane protein insertases/chaperones. Like disruption of the signal recognition particle ( SRP ) co translational pathway, e limination of yidC2 causes loss of ge netic competence stress sensitivity (acid, osmotic and oxidative), and impaired biofilm formation. It has been postulated that a cid sensitivity in these mutants is due to defects in assembly of the F 1 F o ATPase which pumps protons out of the cell during acidic growth conditions Elimination of yidC1 is less severe, with no observable effect on growth or stress sensitivity YidC2 has a longer more positively ch arged C terminal tail than YidC 1 and can complement an Oxa1 mutant in yeast. Oxa1 function depends on its C terminal tail C himeric proteins were constructed to evaluate the roles of the C termini in stress tolerance o f YidC1 and YidC2 Placing the YidC2 tail on YidC1 restored acid and osmotic stress tolerance to a yidC2 mutant In contrast placing the C ter minal tail of YidC1 on to YidC2 resulted in a d ominant negative effect with decr eased growth yield and sensitivity to stress indicating the C terminal domains play an importan t role in stress tolerance.

PAGE 17

17 Studies revealed b oth YidC1 and YidC2 can function in E. coli to assemble an active F 1 F o ATPase in a Y idC depletion strain In S. mutans d eletion of either yidC 1 or yidC2 results in decreased membr ane associated ATPase ac tivity indicating that the pronounced acid sensitive phenotype of the delta y idC2 mutant stems from further mechanism s in addition to impaired proton extrusion. Blue N ative PAGE, which separates native membrane protein complexes, combined with LC MS/MS revealed differences in location of glycolytic enzymes, citrate metaboli c enzymes, transporters, and ribosomal proteins associated with the membrane fraction s of wildtype and the yidC mutant s Therefore YidC1 and YidC2 appear to contribute to assembly of membrane associated complexes involved with glycolysis and to ribosome tethering for co tran slational translocation, likely explaining the dispensability of the SRP pathway in S. mutans Collectively, t hese results provide evidence that the difference in s tress sensitivity between the yidC1 and yidC2 mutant s stem s from functional differences in their C termini

PAGE 18

18 CHAPTER 1 LITERATURE REVIEW Streptococcus mutans and Dental Caries Streptococcus mutans is a member of the v iridans group of streptococci which was originally characterized by partial hemol ysis of red blood cells through the production of hydrogen peroxide, giving blood agar surroun ding colonies a green appearance (alpha hemolysis ) Streptococci are catalase negative, gram positive cocci that form chains. C urrent classification of the viridans group relies on three tests to differentiate them from other streptococci They are leuci ne aminopeptide positive, pyrrol idonylar y lamid a se negative and do no t grow in broth with 6.5% NaCl. The viridians group is further sub di vided into the mutans, salivari us, anginosus, sanguinus and m itis groups, sometimes inaccurately referred to as the oral streptococci as they are found in other lo cations besides the oral cavity [ reviewed in (1 2) ] S. mutans a member of the mutans group, is not capable of hydrolyzing arginine, but produce s a cetoin, hydrolyze s esculin and can metabol ize both mannitol and sorbitol. The species S. mutans is furth er divided into four serotypes; c e f and k based on cell wall rhamnose glucose polysaccharides Most (70 80%) S. mutans strains isolated from the oral cavity are serotype c while ~ 20% are type e and less than 5% are serotype f or k. Serotype k was only recently recognized, when it was isolated from blood of Japanese patient s with bacteremia (3) S. mutans is the bacterium most often associated with human dental caries (4 5) However, it has also been isolated from patients with infective endocarditis and atherosclerosis and a serotype f strain has been shown to invade human coronary artery endothelial cells (6) indicating that certain serotypes of S. mutans may contribute

PAGE 19

19 to cardiovascular disease. In an Immunology Natu re r eview by Taubman et al., 2006 the authors report that The World Oral Health Report of 2003 stated that 60 90% of school child ren and adults have experienced de nt al caries, declaring this infectious disease is a major health problem in industrialized countries (5) In the United States d ental caries is the n umber one childhood disease, and is most prevalent in minority populations affecting ~20 million children from low income families. In 2004 t here was an estimated $75 billion spent on oral health car e alone accounting for ~5% of total healthcare costs in United States As the consumption of refined sugars increase s in developing countries, so does the incidence of dental caries and other diseases associated with poor oral health. S. mutans possesse s numerous virulence factors both s ecreted and membrane associated that contribute to its disease causing properties and to it s ability to contend with the harsh e nvironment of the oral cavity reviewed in (7) S. mutans forms biofilms and is able to attach to th e tooth surface through sucrose dependent and sucrose independent mechanisms both of which require the secretion of several proteins (4) The sucrose dependent mechanisms involve the production of glucans (8) which allow for stable adherence to the t ooth surface, wh ile sucrose independent adhesion involve s the protein P1 (Antigen I/II, PAc) that binds to salivary agglutin in (gp340) on the pellicle of the tooth (9) S. mutans secretes three glucosyltransferase (GTF ) proteins capable of synthesizing extracelluar glucan polymers. These polym ers can be water soluble 1,6 linkages of glucose synthesized by GtfD, or water insoluble with branch ed 1,3 linkages of glucose synthesized by GtfB and GtfC (8) The water insoluble glucan

PAGE 20

20 polymers are known as mutans and are responsible for the formation of dental plaque (8) S. mutans also secretes four glucan binding proteins, Gb pA, GbpB, GbpC, and GbpD, which are important for biofilm formation (8) Within the biofilm, S. mutans is able to transport and metabolize an enormous variety of carbohydrates, conferring much versatility and a competitive advantage over the other species of bacteria (10) I t is through f ermentation of carbohydrates resulting in production of organic acid s that cause s the erosion of tooth enamel which is soluble between pH 5.4 to 4.4 (11) S. mutans is able to cont inue glycolysis at a pH of 4.4 which is beyond the pH of 4.8 where growth stop s (12 13) fur ther lowering the pH in plaque inevitably dissolving the enamel and inhibiting less acid tolerant species of bacteria S. mutans is not only acidogenic; it is aciduric (acid tolerant) and is able to induce an acid tolerance response (ATR) after exposure to sublethal pH conditions The ATR results in differential gene expression, up regulation of stress proteins, increased H+ ATPase activity, and diversion of glycolytic metabolites to less acidic end products, ultimately allowing inc reased survival at what would normally be a lethal pH of 3.0 (14 16) While the glycolytic enzymes of S. mutans are slightly more acid resistant tha n other oral bacteria, it is the cytoplasmic membrane that plays the largest role in acid tolerance (13) In fact, the composition of lipids in the membrane is regulated in response to acid pH. In a study by Fozo et al. 2004 it was found that a fabM mutant, which is inhibited in its ability to produce unsaturated fatty acids had a decreased ability to contend with low pH, decreased glycolytic capability and distorted glucose PTS activity (17) This same group found that the lipid composition of the wild typ e membrane rapidly changed from

PAGE 21

21 short chain saturated membrane fatty acids at pH 7.0 to long chained, monounsaturated fatty acids at pH 5.5 (18) D espite the im portance of the membrane to virulence, in comparison to E. coli, very little is known about the process es involved in its biogenesis or the secretion of proteins through it The major aim of this study was to investigate the roles of two me mbra ne associated chaperone inse rtases, YidC1 and YidC2, in S. mutans membrane biogenesis Acid Tolerance F 1 F o ATP s ynthase S. mutans is a facultative anaerobe and does not contain a cytochr ome oxidase system. Therefore, ATP production is facilitated primarily through substrate level phosphorylation during glycolysis and fermentation producing organic acids (lactate, acetate and formate), which separate into protons and their anionic forms effect ively lowering the pH of the cell The membranes of stre ptococ ci are permeable to protons which can flow back into the cytoplasm and lower the internal pH, reviewed by Quivey et al. 2001 (19) The membrane potential or proton motive force (PMF) is maintained by pumping p rotons from the cytoplasm to outside the cell by the F 1 F o H + ATPase at the expense of ATP, as well as through the transp ort of metabolic end produc ts coupled to proton extrusion (20) The F 1 F o A TPase is composed of two domains The soluble F 1 portion found in the cytoplasm is made up of 3 3 form s the catalytic domain responsible for the ATPase activity The F o integral membrane portion is composed of ab 2 c 10 which forms the proton translocating channel (21) Together these subunits form a functional enzyme, which in E. coli utilizes the electrochemi cal gradient of the membrane to syn thesize ATP I n S. mutans an d other

PAGE 22

22 facultative anaerobes the main function of this multi subunit enzyme is to extrude protons through the hydrolysis of ATP However, work by Sheng and Marquis et al. 2006, showed that t he F ATPase is also capable of brief synthase activity when starved cells are exposed to pH 3.0. This was theorized to provide ATP for concurrent proton extrusion (22) The F 1 F o ATPase is an important factor in the acid tolerance of S. mut ans with a pH optimum of 6.0 but nearly 75% maximal functional activity at pH 5.0 (23 24) It s expression is up regulated in response to low pH, ther eby increasing this or ganism s capacity to tolerate acid end products created through metabolism (12, 25) W ork by Magalhaes, et al 2005 characterized a P type H + ATPase, which the authors propose to function in conjunction with the F 1 F o ATPase to extrude protons aiding in acid tolerance (26) In support of this hypothesis, a recent microarray study found the gene p acL encoding a ca tion transporting P type ATPase, was upregulate d fourfold during acid adaptation in S. mutans (27) The same study demonstrated up regulation o f the F ATPas e operon by about three fold The role of this P type ATPase in acid tolerance needs further exploration Assembly of the F 1 F o ATPase has been investigated in E. coli These studies have revealed that insertion of the F o membrane components of F 1 F o ATP synthase requires several pathways In which contains one transmembrane domain with its N terminus located with in the membrane, requires the signal recognition particle (SRP) pathway the SecYEG translocon and the aid of YidC f or insertion (28) five transmembrane segments, also requires the SRP, SecYEG translocon and YidC for insertion (28 29) which contains two transmembrane domains with its N and C

PAGE 23

23 termini located in the periplasm, is inserted solely by the YidC only pathway in E. coli (30) In S. mutans it is unknown which proteins mediate the insertion of the F 1 F o ATPase, howeve r previous research from the Brady group showed a decrease in membrane associated ATPase activity in mutants of the SRP pathway, and the yidC2 mutant (31) This will be discussed in more detail below. Carbohydrate utilization and d iversion of glycolyti c m etabolites S. mutans and other lactic acid bacter ia in the oral cavity have many methods for dealing with the constant ly changing nutrient availability found in the oral biofilm For example, S mutans has two differen t mechanisms f or transporting carbohydrates including ABC transporters and phosphoenolpyruvate phosphotransferase systems (PEP PTS) T here are four possible ABC transporters involved with sugar transport, including the MSM (multiple sugar metabolism) transporter which transport s sugars at the expense of ATP (32) Some ABC transporters consist of two membrane permeases, two membrane assoc iated ATPases, and a solute binding protein located on the outside of the membrane The other way carbohydrates are transported is through sugar specific p hosphoenolpyruvate sugar p hosphotransferase systems ( PEP PTS), where sugar s are transported and pho sphorylated by the sugar specific EII permease S. mutans is pre dicted to have 14 PTS systems, five of which are constitutively expressed and capable of transport ing glucose, fructose, maltose and sucrose at all times (32) There are also additional genes for PTS systems for transport of lactose, fructose, mannose, cellobiose, trehalose, glucoside s mannitol, sorbitol, rib ulose and sorbose/mannose. Each PTS system is composed of two components, the EI component, which is common to all PTS systems, and t he sugar specific EII component, generally composed

PAGE 24

24 of domains A, B, and C (33) Domains EIIA and B are phosphorylated and in turn transfer the phosphate to the incoming s ugar, which is transported by the membrane integral EIIC domain (33) The phosphocarrier protein HPr and enzyme EI facilitate the transfer of the phosphate from phosphoenolpyruvate (PEP) to the EIIAB components of the PTS systems HPr has two phosphorylation sites at His 15 and Ser 46 (34) Enzyme EI is responsible for phosphorylation of HP r His P, while HPr kinase/phospha tase is responsible for phosphorylation/dephosphorylation of HPr at Ser 46 HPr Ser P is a major compon ent of carbon catabolite repression (CCR) in S. mutans and through the regulation of HPr kinase/phosphatase activity allows for use of preferred carbohydrates over less desired ones, thus maximiz ing the energy production within the cell. M echanisms by which S. mutans regulates CCR are not completely understood and probably have unique components t hat vary depending on t he genes or pathways being regulated (35) However, the ability of S. mutans to perform this regulation over carbohydrate utilizatio n is likely critical to its virulence just as efficient energy production is important for proton extrusion Furthermore, there is a clear association of membrane biogenesis in the process of sugar transport. Once a sugar is transported into the cell and phosphorylated, it ent ers the glycolytic pathway and through substrate level phosphorylation it is metabolized ultimately producing ATP and pyruvate. Pyruvate metabolism is a key branch point in the cell (36) Pyruvate can be used in fermentation reactions where it is converted to lactate by lactate dehydrogenase (producing NAD+ from NADH), which restores t he redox balance of the cell. A lternatively it goes the way of pyruvate formate lyase (PFL) which results in formate, ethanol, and acet ate, producing an additional ATP through the

PAGE 25

25 production of formate The redox balance is also preserved in this set of reactions by bifunctio nal acetaldehyde Co A /alcohol dehydrogenase, which oxidizes two NADH to two NAD+ for every pyruvate processed Whe n sugar concentrations are high pyruvate is directed to the lactate branch through activation of la ctate dehydrogenase by fructose 1 6 bis phosphate. This buildup of toxic intermediates while maintaining the redox balance of the cell (20) The PFL branch is only active during anaerobic conditions as PFL is deactivated by oxygen An alternative branc h, which is implicated in acid tolerance, is oxidation of pyruvate by pyruvate dehydrogenase forming acetyl CoA and CO 2 (36) Acetyl C oA can be further processed by bifunctional acetaldehyde Co A/alcohol dehydrogenase resulting in ethanol, or by acetate kinase producing acetate (36 37) F o rmate, with a pKa of 3.77, is the most acidic end product of S. mutans fermentation compared to lactate (pKa of 3.86) and acet ate ( pKa of 4.76 ) B y r egulating the enzymes involved in fermentation, S. mutans can control acid production In a two dimension al proteomics study by Len, et al. 2004 comparing S. mutans grown in continuous culture at pH 5.0 compared to pH 7.0 changes were limited to three key biochemical pathways involved with acid adaptation ; glycolysis, acid production, and branched chain amino acid synthesis (14) These a uthors concluded that S. mutans diverts glycolysis toward less acidic end products such as lactate and ethanol and diverts pyruvate to branch chain ed amino acid synthesis Branch chaine d amino acid biosynthesis combats low pH by consuming the reducing equivalents pyruvate, 2 oxobutanoate, and NADPH in effect preventing them from being used in pathw ays that produce acid. Additionally the production of

PAGE 26

26 branch chained amino acids results i n NH 3 which reacts with H+ to form NH 4 + effectively increasing the internal pH of the cell (14 15) Two component s ystems and acid tolerance Unlike Escherichia coli and Bacillus subtilis which co ntain a large number of alternative sigma factors that regulate gene expression in response to stress, S. mutans only contains two, 70 and x (7) Consequently two component signal transduction systems (TC S or TCSTS ) are very important for sensing environmental signals and enacting stress tolerance in S. mutans All bacteria encode genes for t wo component system s, reviewed in (38) In general, they contain a membrane bound sensor histidine kinase (HK) that senses a signal resulting in auto phosphorylation followed by transfer of the phosphate to a soluble cytoplasmic response regulator (RR) This changs the confo rmation of the RR, which then positively or negatively regulates transcription of a target gene (38) The S. mutans genome is predicted to encode as many as fourteen TCS s depending on the strain that a ffect expression of virulence factors, biofilm formation, competence development, stress tolerance and bacteriocin production (7, 39) A whole genome transcriptional analysis of S. mutans in response to acid adaptati on identified a nu mber of TCS s as being up regulated in response to pH 5.5 conditions These included CiaHR, LevSR, LiaSR, ScnKR, Hk1037/Rr1038, and ComDE (27) Another study showed that VicK of the VicR/K TCS, is also involved in acid tolerance in S. mutans (40) Many of these TC S s overlap in stress response pathways and it i s believed there is cross talk between the different TCS s (41 42) A recent study by Banu et. al. 2010, showed pknB and pppL which en code a serine/threonine kinase and phosphatas e respectively, are capable of affecting pathways regulated by the V icRK and ComDE TCS s (43) The

PAGE 27

27 Lia SR is a TCS involved with adaptation to cell envelop stress, with a n intramembrane sensing histidin e kinase that responds to cell wall envelope damage rather than a stress signal (44) A study by Senadh eera et al. 2009, found that SMU .1727 ( yidC2 ), has a CesR/LiaR binding moti f upstream of its promoter and is induced under conditions of envelope stress, brought on by the lipid II cycle inhibitor ba citracin (45) Other lipid II cycle inhibitors, such as vancomycin and nisin, also induce the LiaSR TCS, activating genes involved in cell wall peptidoglycan synthesis and membrane prot ein biosynthesis. Using qrt PCR a number of other proteins involved with membrane biogenesis were also identified as targets including; ftsY ropA (trigger factor), ftsH and degP ( htrA ) Also, a hy pothetical membrane protein (SMU .753) containing a PspC ( phage shock protein C) domain, was up regulated tenfold during envelop stress conditions. Secretion and Membrane Protein Biogenesis Escherichia coli. The bacterial general secretion pathway and membrane protein insertion has been most extensively studied in E. coli Development of techniques to purify membrane proteins involved in secretion and their reconstitution into in vitro systems of proteoliposomes have allowed researchers to reconstruct and evaluate minimum requirements for tr anslocation of model substrates. Also, development of conditional expression systems of essential components has allowed researcher s to evaluate roles of individual components in the secretion process in vivo Through this work a model of protein secreti on has been developed for E. coli and recently reviewed in (46 48) Targeting begins at the ribosome f or proteins destine d for secretion or membrane insertion Sec reted proteins contain an amino terminal signal sequence com posed of a positively charged amino ( N) domain a hydrophobic ( H) domain and polar carboxy terminal ( C) domain, which is recognized by trigger factor, a chaperone

PAGE 28

28 with peptidyl prolyl cis trans i somerase activity (49) Both t rigger factor and the signal recognition particle (SRP) bind to ribosomal protein L23 and compete for bin ding to signal sequences of nascent polypeptide chains as they exit the ribosome (50) Membrane proteins contain a n N terminal signal anchor sequence that is highly hydrophobic and is preferentially bound by the SRP and targeted for co translational translocation Alternatively, p roteins recognized by t rigger factor are bound by the SecB chaperone and targeted post transla tional ly to the SecA molecular motor ATPase protein, which is associat ed with the SecYEG translocon and facilita tes the translocation of secretory proteins through the membrane As the pre protein is secreted the signal seque nce is cleaved by signal peptidase resulting in a mature protein In E. coli the essential universally conserved SRP pathway is compo sed of Ffh 4.5S sc RNA (114 nt) and the FtsY receptor. The eukaryot ic SRP pathway, which targets proteins to the endoplasmic reticulum, is more complicated with a dditional proteins associate d with the particle However, it does have S RP54 (Ffh homolog), a 7S RNA and a receptor composed of SR subunit is homolog o us to FtsY (46) Ffh and the scRNA (small cytoplasmic RNA) form a complex, which recognizes hydrophobic signal sequences of nascent polypeptide chains as they exit the ribo some and targets the ribosome nascent chain (RNC) to the membrane receptor FtsY Both Ffh and FtsY are GTPases, and upon binding to one another undergo a conformational change, resulting in binding subsequent hydrolysis of GTP This allows the SRP pa rticle to transfer the RNC to the SecYEG translocon pore where the membrane protein is inserted co translationally SecY and Sec E are essential in E. coli and are homolog o us to the eukaryotic Sec61 proteins. SecYEG associates with

PAGE 29

29 a number of accessory proteins including SecDF(Y ajC ) and YidC SecDF(Y ajC ) are believed to aid in protein translocation and may regulate th e cycling of SecA during translocation (51) SecA is required f or insertion of membrane proteins with large hydrophilic segments. Mutants in SecD and F exhibit a cold sensitive phenotype and deletions of SecDF results in extremely slow growth with defects in protein secretion (52 53) The YidC protein functions i n both Sec dependent and Sec in dependent pathways and will be discussed in greater detail below. Gram positive bacteria B. subtilis has been us ed as the model organism for secretion studies in gram positive bacteria; however most studies have focused on secretion, with little work done on membrane biogenesis. Similar to E. coli B. subtilis has homologs for SecYEG, SecA and SecDF (Y ajC ), as well as the SRP pathway. However, there is no SecB homolog and SecDF is one protein (54) instead of two separate ones as in gram negative species In B. subtilis SecDF mutants exhibit a cold sensitive phenotype and a diminished capacity to secrete proteins in high volume (55) SecDF is absent from the Streptococci, but is found in S taphylococci, while YajC seems to be universally c onserved in gram positive bacteria. T he SecE prote in of B. subtilis and other gram positive bacteria are smaller than the E. coli SecE and contain only one TMD where the E. coli SecE has three TMDs. However, only the C terminal TMD and cytoplasmic domain are essential for SecE function in E. coli (56) T his r egion is conserved in the SecE proteins from gram positive bacteria (46) SecG is non essential in B. subtilis but deletion results in secretion defects and cold sensitivity. Another difference from E. coli is that the SRP of B. subtilis contains an additional protein HBsu, with no n specific DNA binding activity. T he B. subtilis scRNA is 270 nucleotides with an

PAGE 30

30 Alu domain that is also found in scRNA of Clostridium perfringens and Listeria monocytogenium (55, 57) but not S. mutans (31) The SRP pathway is essential for grow t h in B. subtilis and depletion of Ffh results in an altered cell morphology and a defect in protein secretion (58) The contribution of the SRP to protein secretion in E. coli is less apparent When hybrid SecYEG translocons from E. coli and B. subtilis were evaluated for translocation in an E. coli backgrou n d SecA from B. subtilis bound with low affinity to SecYEG compared to SecA from E. coli regardless of the source of SecYEG (59) Additionally this study found that while hybrid translocons were stable, t hey were inefficient at protein translocation. A surprising finding in S. mutans was that the SRP pathway is dispensable for viability (60 61) M utants in Ffh, scRNA or FtsY are however, impaired in environmental stress tolerance (31) The SRP pathway is also dispensable in S. pyogenes where it is required for virulence (62) The ability of streptococci to survive wi thout an SRP pathway is likely related to the presence of two yidC homologs, on e with a long er positively charged C terminal tail, which will be discussed belo w. A number of gram positive species have accessory Sec proteins dedicated to secretion of a subset of proteins often involved with virulence (63) In some species of Streptococcus ( S. gordo nii S. pneumoniae S. parasanguinis) S taphylococcus ( S. aureus S. hemolyticus S. epidemidis ) and Bacillus ( B. cereus B. anthracis B. thuringiensis ) the accessory Sec system incl udes accessory SecA 2 and SecY 2 protein s which function separately from the canonical SecA and SecY proteins of the general secretion pathway. The accessory Sec locus of S. gordonii is located downstream of its substrate GspB, a large ser ine rich cell surface glyco protein that

PAGE 31

31 binds platelets (64) Located in the same locus are several other genes required for and function. These include tw o proteins with homology to Sec E and SecG and proteins n eeded for the glyco s ylation of GspB, which is req uired for its function (65) There is a similar gene organization in the accessory Sec locus in S. parasanguinis with surface protein Fa p 1, followed by the accessory SecA2/Y2 genes and genes r equired for glycosylation of Fap 1 (66) This is also true for S. aureus S. epidermidis S. agalactiae and S. pneumoniae of which all encode serine rich repeat (Srr) proteins involved with adhesion (66) A number of other gram positive species, Mycobacterium, Listeria, and Corynebacterium, encode only a SecA2 gene, and there is no conservation in the gene locus of these species (63) An other interesting finding in the S treptococci was the discovery by the Caparon group of the Exportal in S. pyogenes a micro domain containing SecA and dedicated to secretion (67 68) A similar micro domain was identified in S. mutans with SecA and Sortase A found to c o localize at a distinct site in the membrane (6 9) However, another group using a similar technique, found SecA to be distributed throughout the cell in S. pyogenes (70) An E xportal has not been ide ntified in Bacillus, where the Sec translocon is found in several locations in a spiral arra ngement along the cell (71) It is clear there are differences in protein secretion pathway s and membrane biogenesis among bacterial species. Gram negative bacteria contain two cell membranes, with a periplasmic compartment in between, while gram positive bacteria contain one membrane, but have a thick peptidoglycan cell wall. Much work remains to be done in gram positive bacteria with regard to protein secretion mechanisms, including the S treptococci. It is pos sible there are as yet unidentified accessory proteins

PAGE 32

32 or pathways involved in protein secretion and membrane biogenesis in this important genus of bacteria. Stress Response and Membrane B iogenesis in S mutans Co translational signal recognition particle p athway Previously it was believed the SRP pathway was essential for viability in all cells including bacteria (58, 72) H owever a search for genes involved in acid tolerance in S. mutans using transposon mutagenesis yielded a mutan t with a disruption of ffh (60) It was later demonstrated that an acid adapted isogenic mutant of ffh had decreased membrane associated ATPase activity, while the ATPase activity of permeabilized whole cells was unaffected when compared to the wild type suggesting a def ect in the assembly at the membrane in these mutants (73 74) L ater it was confirmed tha t elimination of the entire SRP pathway of S. mutans is not lethal but results in an inability to tolerate environmental stresses (acid, osmotic and oxidative), decreased biofilm formation, and loss of natural competence (31, 75) To investigate the involvement of the SRP pathway in acid tolerance, 2D gel electrophoresis of membrane pro teins was performed using cells of SRP mutants grown at pH 7.0 or pH 5.0 and compared to the wild type cells grown under th e same c onditions. Result s showed an increase in the molecular chaperones DnaK, GroES and the ClpP protease a s well as a decrease in a number of metabolic enzymes The subunit of the F 1 F o ATPase was also decreased in the Ffh membrane preparations Transcriptome analysis was a lso performed comparing the Ffh mutant and wild type under non stress conditions. Results were consistent with a global stress response showing an increase in molecular chaperones and proteases, most likely caused by defects in membrane protein biogenesis (75) There was also an increase in genes in vol ved with detoxification, including a number of oxidoreductases.

PAGE 33

33 Consistent with the decrease in competence and aberrant biofilm formation seen in the ffh mutant, th ere were a number of genes down regulated in the competence pathway (75 76) Since the SRP co trans lational pathway is essential in other bacteria there must be another mechanism for co translational translocation in S. mutans In mitochondria co translational translocation is mediated by Oxa1, which is able to bind mitochondrial ribo somes by way of its positively charged C terminal tail Analysis of the S. mutans genome revealed there are two homologs of the YidC/Oxa1/Alb3 family YidC1 and YidC2 Elimination of y idC2 results in a stress sensitive phenotype similar to the SRP pathwa y mutants, with growth impairment under acid, osmotic and oxidative stress conditions, decreased membrane associated ATPase activity, decreased genetic competence, and impaired biofilm formation (31) Disruption of y idC1 has a much less severe effect, with no obvious growth defects or stress sensitivity The y idC1 mutant does however, display aberrant biofilm formation Attempts to isolate do uble mutants in the SRP pathway and YidC2 have not been possible. Nor is it possible to eliminate both Y idC1 and YidC2 simultaneously, suggesting functional redundancies in the SRP and YidC2 pathways, as well as between YidC1 and YidC2 in S. mutans Yid C/ Oxa1 Alb3 Family of Proteins The Oxa1/YidC /Alb3 family of proteins is universall y conserved in all three domains of life (77) Oxa1 o f th e inner mitochondrial membrane YidC in the bacte rial cytoplasmic membrane, and Alb3 in the thyla koid membrane of chloroplasts all possess conserved functions in insertion of respiratory chain complexes, such as the cytochrome oxidase systems, F 1 F o ATP synthases and light harvesting chlorophyll binding proteins in plants (78 80) There are common structural features among this

PAGE 34

34 family of proteins, with the highest sequence conservation in the 5 C terminal transmembrane domains (81) Experiments have shown cross species complementation is possible am ong family members, indicating vestigial functions still remain Oxa1 of m itochondria The mitochondrial YidC homolog, Oxa1, is located in the inner mitocho ndrial membrane and is the founding member of the Oxa1/ YidC/Alb3 family of proteins Oxa1 was discovered in a yeast mutant that lacked critical components of the cytochrome c oxidase complex and was respir ation deficient, hence the name Oxa for ox idase a ssembly (82) Because of its highly basic C terminal tail which has been shown to interact with mitochondrial ribosomes Oxa1 is capable of co translational insertion of mitochondrial enco ded proteins There are no SRP or SecYEG homologs in the mitochondrial inner membrane I nstead these functions are filled by Oxa1, which is also capable of post tra nslational insertion of nuclear encoded proteins that are first imported into the matrix t hrough the TOM (translocon of outer membrane) and TIM (translocon of inner membrane) complexes located in the outer and inner membranes of the mitochondria (83 85) Oxa1 is involved with the insertion of subunit 9 of E. coli ) of the F 1 F o ATPase, CoxII of the cytochrome oxidase comp lex and Oxa1 it self in Saccharomyces cerevisiae (84) S. cerevisiae contains another Oxa homolog, Cox18/Oxa2 which is also involved with cytochrome oxidase assembl y (86) It is now apparent, throug h the work of Funes et al (87) that Cox18/Ox a2 makes up a second branch of the Oxa1/Y idC/Alb3 family specifically involved with the biogenesis of the cytochrome c oxidase complex, whereas Oxa1 probably functions in a more general way for insertion of inner membrane proteins (87

PAGE 35

35 88) Oxa1 and Oxa2/Cox18 are universally conserved in mitochondria of plants, fungi, and animals and have been shown to function to varying extents in heterologous species (87 88) YidC of E coli The YidC protein was discovered in E. coli when proteins that were previously thought to insert spontaneously were impaired in insertion in a y idC depletion strain (89) YidC from E. coli contains six transmembrane segments (TMs) and a large periplasmic loop between TM1 and TM2. This topology is common among YidC proteins of gram negative bacteria. While the five C terminal transmembrane domains are highly conserved, the periplasmic loop is varia ble in both length and sequence (90) The first transmembrane domain of YidC serves as an un cleaved signal ancho r and is not vital for function (91) N or is the periplasmic loop, which even with a deletion from amino acid 25 323, does not affect Yid C function (91) Purified YidC has been shown to form both monomers and dimers (92) Furthermore, YidC has both sec dependent and sec independent functions. C onsistent with this finding, YidC is expressed i n excess of t he SecYE translocon, with 2,700 copies per cell compared to 100 200 respectively (93) It has been shown that i nsertion of YidC it self require s the SRP SecA and SecYEG YidC pathways (93) YidC is essential in E. coli and the generati on of a yidC depletion strain JS7131 where expression of YidC was placed under the control of the AraBAD promoter, has m ade many functional studies of Y idC possible (89) A number of YidC substrates have been identified, leading to a better understanding of YidC function. CyoA of the cytochrome bo3 requires both SecYEG and YidC for proper insertion (94) while subunit of the F 1 F o ATPase, M13 and pf3 phage coat proteins are subs trates of the

PAGE 36

36 YidC only pathway (95 96) MscL, the mechanosensitive channel of large conductance, was shown to require both the SRP and YidC for proper membr ane insertion, but did not need SecY EG, suggesting the SRP may target proteins to YidC for insertion (97 98) There is also evidence for YidC as a chaperone involved in the assembly of polytopic membrane proteins such as MalF (99) LacY (100) and MltA (101) Based on these experiments it i s postulated that YidC is involved with release of transmembrane domains from the translocon and possibly in the ass embly of multimeric protein complex es Alb3 of c hloroplasts I n c hloroplasts of Arabidopsis thaliana Alb3 is essential for viability (102) It is involved with the biogenesis and insertion of the LHCP (light harvestin g chlorophyll binding proteins) complex and was first discovered in chloroplasts with an albino phenotype (102 103) W ork by Gerdes et al. (104) showe d the existence of a second YidC/Oxa1 /Alb3 homolog in Arabidopsis thaliana termed Alb4 located in the thyla koid membrane (104) However, since only Alb3 is essential, the two homologs must have somewhat different functions. There are also two homologs in the unicellular algae C hlamydomonas reinhardtii termed Alb3.1 and Alb3.2, with Alb3.2 being essential for viab ility (105) Mutations in Alb3 .1 indicate that it functions in the insertion of LHC proteins, where reduction of Alb3.2 through RNAi indicate it s involved more s pecifically with photosystems I (PS1) and II (PSII) assembly wit h little effect on LHC p roteins. The thyla koid membrane is different from the mitochondria l inner membrane where Oxa 1 and Oxa2 are located. The thyl a koid membrane of chloroplast contain homologs of SecA, SecY and Sec E, as well as an SRP pathway reviewed in (106) Additionally, thyla koids possess a Tat pathway for transport of a subset of

PAGE 37

37 proteins including folded proteins that is homologous to the bacterial Tat system (107) In chloroplasts, the cpSRP is in volve d in both a co translational pathway that targets proteins to the cpSecYE complex (108) and a post translational pathway that targets to the Alb3 translocase. The cpSRP is composed of the conse rved cpSRP54 (Ffh homolog) and cpFtsY receptor. It differs from other SRP pathways in that it does not contain an RNA compon ent, and has an additional unique protein component, cpSRP43 that is necessary for post translational targeting of cpSRP substrates to Alb3 Recent work has shown that the C terminal tail of Alb3 is invol ved with targeting of the cpSRP to the thylokoid membrane and cpFtsY through an interaction with cpSRP43 (109 110) Alb4 also has a long C terminal tail with similarities to that Alb3 but does not interact with cpSRP43, and instead is proposed to react with SecYE or with ribosomes (110) There is also evidence that Alb3 f orms stable complexes with the cpSecYE complex (111) The Alb4 protein was shown to be involved in assembly of C F 1 C F o ATPase by stabilizing the interaction between the CF 1 to the CF 0 (112) Alb3 was not involved in this process. YidC1 and YidC2 in gram p ositive bacteria While gram negative bacteria have only one YidC, m any g ram positive bacteria contain two genes encoding proteins of the O xa1/YidC/Alb3 family (77) However, n ot all gram positi v e bacteria encode two homologs. For example, Staphylococcus aureus encodes only one YidC, while the other Bacillales enco de two In general, g ram positive species in the order of Lactobacillales have two YidC homologs and one is usually shorter than the other by roughly 35 amino acids (81) Such is the case for S. mutans with YidC1 containing 271 ami no acids and YidC2 containing 310, with the main difference being in the length of

PAGE 38

38 the C terminal tails. Many gram positive homologs are predicted to be lipoproteins that are processed by signal peptidase II (113) Based on a common location within the genome it appears that yidC1 is more closely related to s poIIIj from B. subtilis and y idC fro m E. coli which are all located downstream from rpnA the gene encoding ribonuclease P In gram positive species this locus also contains an RNA binding protein of the Jag family located downstream from the yidC1 gene This gene arra ngement is conserved a mong the S treptococci and B. subtillis In S. mutans yidC1 appears to be located in an operon with Ribonuclease P and the Jag RNA binding protein, while the E. coli yidC in not in an operon (80) The location of the YidC2 homologs is conserved in the S treptococci, with the gene for acyl phosphatase ( acp ) upstream from yidC2 H owever this arrangement in not observed for yqjG the second YidC homolog of B. subtilis. When comparing the conserved five C terminal transmembrane domains of the YidC/Oxa1Alb3 family using PSI BLAST there is roughly 20 to 30% se quence identity and approximately 40 to 50% similarity between the homologs There is a higher level of homology between closely related species. For example the sequence similarity between the S. mutan s YidC proteins and those from S pyogenes is 64% i dentity and 79% similarity between YidC1 and Spy1 (275 a.a.) and 58% identity and 75% similarity between YidC2 and Spy2 (307 a.a.). When comparing the sequence of S. mutans YidC1 to that of YidC2 there is only 30% identity a nd 50% similarity between them suggesting an early gene duplication event and evolution of two separate proteins. There is relatively little i n the literature about the YidC proteins from gram positive species compared to E. coli YidC. Most of the studies so far have involved the B.

PAGE 39

39 subtilis homologs, S poIIIj and YqjG, and the two para logs from S. mutans YidC1 and YidC2. B. subtilis SpoIIIJ and YqjG do not display the same differential in length between them as seen for streptococcal homologs SpoIIIJ contains 2 61 amino acid s and YqjG, 275 amino acids (81) SpoIIIJ was first identified as a pr otein essential for spore formation while YqjG is not required for this process (114) E limination of one or the other has very little affect on normal vegetative growth but at least one of these two proteins must be present for B. subtilis to grow, suggesting that overlap sustainably in function (114 115) It was also shown that SpoIIIJ and YqjG function in post translocational folding of secreted proteins (115) which is a somewhat novel function, since in E. coli elimination of yidC has only a minor a ffect on secreted proteins (89) Recent work by Saller et al. 2009 found that YqjG and SpoIIIj are involved in membrane biogenesis W hen over expressed in B. subtilis or in E. coli both co puried with the entire F 1 F o ATPase (116) This study also showed that either SpoIIIJ or YqjG could functionally complement a yidC depletion strain in E. coli Another stu dy by Saller et al. 2011, showed that YqjG is involved in genetic competence development in B. subtilis and through a conditional expression system showed that when spolllj is depleted in a yqjG strain the Lia SR envelope stress response is induced (117) Yqj G is only expressed when SpoIIIJ is absent (118) and work by Chiba et al. 2009 showed the presence of a cis acting r ibosome na s cent chain sensor MifM/YqzJ located directly upstream from YqjG/YidC2, that allows transcription of Y q jG in the absence of SpoIIIJ (119) This regulation is similar to the SecM system that is activated in response to a secretion defect, resulting in increased SecA levels (120)

PAGE 40

40 S. mutans YidC1 and YidC2 Most of the available information regarding YidC1 and YidC2 has come from research from the Brady Lab and those of our collaborators. As mentioned above, elimination of yidC2 results in a similar stress sensitive phenotype as the SRP mutants, with an inab ility to grow when confronted with acid, salt or oxidative stress (31) The yidC2 mutant also displays a pronounced lack of genetic competence, and a defect in biofilm form ation (31) A C terminal deletion mutant of YidC2 displays an intermediate stress sen sitive phenotype with a minor de crease in growth rate during non stress conditi ons and a more pronounced affect under acid and salt stress conditions. Elimination of yidC1 has almost n o influence on growth, stress tolerance or genet ic competence, but does a ffect biofilm formation under conditions of acid stress (unpublished data) Additionally, elimination of yidC1 or y idC2 a ffect s surface adhesin P1 but in different ways (unpublished data, J. Brady and P. Crowley). In the yidC2 m utant, the maturation of P1 is a ffected, as evidence d by reduced immunorea ctivity of certain, but not all monoclonal antibodies that recognize epitopes important for the function of P1 (121) This conclusion was supported by the fact that cells from the yidC2 mu tant exhibit substantially less binding to salivary agglutinin, as measured by whole cell BIAcore surface plasmon resonance assay (122) Disruption of yidC1 resulted in increased immunoreactiv i ty by all monoclonal antibodies and polyclonal antiserum against P1, and increased binding of whole cells to salivary agglutinin This suggests either there is more P1 on the surface of the yidC1 mutant or les s of other protein s resulting in increased exposure of P1 to the antibodies. Cros s s pecies complementation studies between S. mutans and Saccharomyces cerevisiae revealed that YidC2 could mediate co translational translocation in an Ox a1 mutant

PAGE 41

41 (123) This ability was dependent on the presence of YidC 2 C terminal tail The ability of YidC1 to substitute for Oxa1 could not be assessed since it was not properly inserted into the mitochondr ial membrane. Additionally it w as shown that like Oxa1, full length YidC2 interacts with yeast ribosomes When the re verse experiment was done both yeast homologs Oxa1 and Cox18 were able to restore growth to a S. mutans yidC2 mutant under stress cond itions, with Oxa1 showing be tter growth restoration tha n Cox18. Additionally, Oxa1 but not Cox18, fully restore d membrane associated ATPase activity and partially restored genetic competence to a yidC2 mutant This is consistent with the fact that S. mutans ATPase activity is similarly diminished in the yidC2 C terminal deletion mutant as it is in the complete yidC2 mutant strain, indicating that the C terminal tail is important for ATPase assembly. The C terminal tail of Cox18/Oxa2 is shorter than th at of Oxa1, and was not able to function in either of these capacities. On the other hand, genetic competence is severely affe cted by a complete deletion of y idC2 and o nly partially by elimination of the YidC2 C terminal tail T herefore there are functi ons of YidC2 that do and do not rely on the C terminal tail, thus explaining the partial complementation of genetic competence by Oxa1 Recent w ork by Suntharalingam et al. 2009 found that the promoter region of S. mutans yidC2 contains a consensus bind ing s ite for LiaR, of the Lia SR two component system involved with sensing envelop stress signals (45) They showed u sing qRT PCR that yidC2 expression is up regulated three fold by L ia SR during the envelope stress response ( induced by exposure to bacitracin ) This may be homologous to the phage shock response in E. coli that is turned on during YidC depletion (124) R ecent

PAGE 42

42 findings by Saller et al. 2011 (117) discu ssed abo ve, found induction of the LiaSR system in B. subtillis upon depletion of both SpoIIlJ and YqjG (117) YidC1 and YidC2 from S. mutans are predicted to be lipoproteins processed by SPaseII based on the presence of a consensus se quence in their signal peptides Within the lipoprotein processing signal of the yidC1 gene from Enterococcus faecalis is a pheromone peptide involved in induction of conjugative transfer of a pheromone inducible plasmid pCF10 which encodes antibiotic resistance genes (125) This sequence is not present in the signal sequence of either yidC1 or yidC2 from S. muta ns Summary and Specific Aims Streptococcus mutans is an important pathogen affecting 90% of the world population. Dental cavities a re so ubiquitous affecting nearly everyone at some point in their life that m it is a preventable infectious disease In areas of low socioeconomic status, many people cannot afford to see a dentist, and in extreme cases this disease ca n progress to a point where all teeth are lost or can result in systemic infections (5) B illions of dollars are spent each year in the United States alone on the tr eatment of tooth decay, largely the result of the acidogenic and aciduric properties of S. mutans In this study two S. mutans membrane proteins YidC1 an d YidC2, of the universally conserved YidC/Oxa 1 /Alb 3 family of chaperone insertases, were investigated. The inf ormation gained from this research could p otentially be applied to other S treptococci, leading to a greater under standing of the mechanisms of p athogenesis that invo lve integral membrane and secreted proteins. This could ultimately result in more successful tr eatments and potentially prevention of this medically important genus.

PAGE 43

43 Specific A im 1: Develop Tools to Examine Compensatory or Redundant Functions in Membrane B iogenesis of S mutans. S mutans appears to have ov erlapping functions between the SRP c o translational translocation pathway and YidC2. It is not yet understood why there are tw o YidC homologs in gram positive bacteria. This aim st rives to demonstrate that the two Y idC proteins h ave separate distinct functions. For this purpose a conditional expression system was developed, where by y idC2 was placed under the control of an induc ible promoter and the yidC1 gene was eliminated. Also two chimeric proteins wer e made to investigate functional differences in the C terminal domains of YidC1 and YidC2, and their respective roles in stress tolerance. Also a number of synthetic peptide s were designed and used to produce antisera against YidC1, YidC2 and SecY. Specific Aim 2: Evaluation of YidC1 and YidC 2 s Involvement in the A ssembly of the F 1 F o ATP S ynthase Acid tolerance has been largely attributed to the activity of the proton translocating F 1 F o ATPase in S mutans Previous work in the Brady lab found that ffh and yidC2 mutants are sensitive to acid, oxidative and osmotic environmental stress (31) Also, i t was found that these mutants had decreased membrane a ssociated ATPase activity, although the yidC1 mutant did not. Studies presented here will show that both yidC1 and yidC2 can function to insert the F 1 F o ATPase in E. coli and that deletion of yidC1 in S. mutans can also result in decreased membrane associa ted ATPase activity. This suggests that the acid sensitive phenotype of the ffh and yidC2 mutants is not entirely due to decreased F 1 F o ATPase activity.

PAGE 44

44 Specific Aim 3 : Determine Differences in Membrane Protein Complexes between Wild yidC1 yidC2 Mutants using Blue Native Polyacrylamide G e l E lectrophoresis. Membrane proteins play a critical role in stress tolerance and environmental adaptation in S. mutans This organism can surv ive a wide range of environmental conditions, and challenges in part by employing TCS transduction systems that result in differential gene expression, changes in cell wall biogenesis, and metabolic pathways Clearly, m any of these processes take place at the membrane. This specific aim will address the difference in membrane protein complex composition between wild typ e and yidC mutants of S. mutans using Blue Native polyacrylamide gel electrophoresis This method allows the separation of membrane protein complexes under non denaturing conditions, so that differences between wildtype and mutant membrane s can be analyzed by LC MS/MS (liquid chromatography mass spectrometry/mass spectrometry). This technique was al so combined with 2 dimensional SDS PAGE and Western blot to locate protein complexes containing YidC1, YidC2 and SecY, whose associations within the translocation machinery are poorly understood in S. mutans

PAGE 45

45 CHAPTER 2 DEVELOP TOOLS TO EXA MINE COMPENSATORY OR REDUNDANT FUNCTIONS IN MEMBRANE BIOGENES IS OF S treptococcus mutans Rationale for Study A major difference between YidC1 and YidC2 is the length of their C terminal tails. YidC2 has a longer more positively charged tail than YidC1 (Figure 2 1). To evaluate the functions of these tails chimeric proteins were constructed between YidC1 and YidC2 where by the C terminal domains were swapped (Figure 2 2 ). These proteins were then evaluated for their ability to restore acid and osmotic stress tolerance to a yidC2 mutant by growth curve analysis in broth The chimeric proteins were also evaluated for the ability to grow under acid stress conditions on solid media. The extensive understanding of gene regulation in E. coli has made many conditional express ion systems possible and has enabled studies of many essential proteins. The available systems include inducible promoters ( tetA trp and T7) that a llow over expression of proteins, facilitating their purification O ther systems provide for tight regulation (araBAD promoter), enabling essential proteins to be depl eted slowly, so that their effects can be studied [ reviewed in (126) ] T he arabinose inducible/g lucose repressible promoter araBAD, was used in E. coli to conditionally express yidC in strain JS7131 (89) s ecE in strain CM124 (results in decrease d levels of SecY) (127) f fh in strain WAM121 (128) and f tsY in strain FJP10 (129) In studies in B subtilis the P spac promoter (130) a hybrid between a Bacillus phage promoter SP0 1 and the lac p romoter of E. coli (regulated by IPTG via LacI ) as well as the xylA promoter (inducible by xylose via XylR ), have been successfully used to conditionally express proteins (117) Attempts to use these pro moters in streptococci for controlled expression have failed because they were either poorly inducible or very leaky (131)

PAGE 46

46 Use of the NICE ( n isin i nducible c ontrolled e xpression) sy s tem in Lactococcus latis has been promising. Thi s system uses plasmid based nisRK that encode a TCS system, which induces expression from the P nisA promoter in the presence of nisin (132) This syste m was shown to work in several S treptococcal species ( S. pyogenes S. agalactiae and S. pneumoniae ) but showed some leaky expression in S. agalactiae and S. pneumoniae with very little induction in the presence of nisin. On the other hand the system worked very well in S. pyogenes with very little background expression and 59 fold induction in the presence of nisin (131) One of the draw backs to this system is that it uses a plasmid to express the necessary control elements which are not native to streptococci Also the leaky expre ssion in some of the streptococc i indicates that tight regulation may not be possible in S. mutans It is for this reason a conditional expression system that includes endogenous promoters expressed from the chromosome is needed. The cellobiose oper on of S. mutans characterized by Dr. Lin Zeng from the Burne group at the University of Florida, is tightly regulated through carbon catabolite repression (CCR) (133) There are two promoters in this operon, PcelA and PcelB These were used to create promoter fusions to yidC2 in S. mutans The yidC2 promoter fusions were integrated into the gtfA locus of the S. mutans strain NG8 chromosome and evaluated for their ab ility to control expression of y idC2 In addition yidC1 was eliminated by allelic exchange mutagenesis using a spectinomycin marker to generate a strain that under controlled conditions produces little Y idC2 and no YidC This yidC depletion strain was evaluated for difference s in protein expression profiles using cell fractionation and one dimensional gradient SDS PAGE.

PAGE 47

47 Materials and Met hods Synthetic Peptide Antiserum Synthetic p eptides corresponding to C terminal sequences of YidC1 ( LEDEARE LEAKKRRAKKKAHKKRK ) and YidC2 (NPPKPFKSNARKDITPQANNDKKL ITS) were designed based on membrane topology predictions using TMPred (Figures 2 3 and 2 4) These peptides were synthesized and sent to Proteint ech Group, Inc (Chicago, Illinois) for immunization of two rabbits each The resulting antisera were affinity purified (see Affinity Purification P rotocol below). In S. mutans SecY is predicted to be a 47,688 Da protein with 10 transmembrane domains (TMPred topology prediction program). To make antibodies against SecY, four synthetic peptides wer e designed that corresponded to the extracellular loop betwee n TMD 1 2 (PGINAKSLEQLSKLPFLNML) the cytoplas mic loops between TMD 6 7 (QAEYKIPIQYKLAOGAPTN) and TMD 8 9 (VNPEKTAENLQKNASYIPSV) and to the C terminal tail (GMKQLEQYLLKKKYVGFMNV) P eptides were designed based on topology predictions combined with sequence alignments between S. mutans SecY and SecY from Methanococcus jannaschii for which the structure has been solved (134) All four peptides were combined and used to immunize two rabbits (Proteintech Group, Inc.) The resulting antisera were poorly reactive with too much non specific background reactivity to be useful. Therefore the anti peptide antibodies were affinity purified from antisera (see Affinity Purification P rotocol below) before use in future experiments Antiserum against a synthetic peptide corresponding to a non C terminal p ortion of YidC2 wa s also made T wo synthetic peptides were designed o ne corresponding to the cytoplasmic loop between TMD 2 and 3 (SEKMAYLKP VEDPIQERMKN) and one to

PAGE 48

48 an extracellular loop between TMD 3 and 4 (ALYISTRYTRYTKGIASILGI) (see F igure 2 4 for peptide location s ) E ach peptide was used to immunize a different rabbit. In the interest of time affinity purifi cation was completed by Proteint ech Group, Inc. for a fractio n of the result ing anti sera and sent with the final bleeds. Only one of the antibody preparations made against the cytoplasmi c loop between TMD 2 and 3 ( YidC2 TM 2/3) was reactive with YidC2, and therefore was the one used in future experiments. Affinity p urification The procedure used to affinity purify antibodies used in this study was adapted f rom two protocols, one provided by Proteintech Group, and the other from GE Healthcare, the manufacturer of CNBr activated Sepharose 4B beads (Instructions : 71 7686 00 AD) To make the affinity matrix, 5 mg of total peptide was dissolved in 5 ml Coupling B uffer (100 mM NaHCO 3 500 mM NaCl, pH 8.0 8.3). Next 500 mg of CNBr activated Sepharose 4B (Pharmacia) was hydrated in ice cold 1 mM HC l (pH 3.0) solution on ice for 10 min utes The beads were then placed on a porcelain funnel with a vacuum and washed w it h 250 ml of iced cold 1 mM HCl (pH 3.0) solution The Sepharose b eads were not allowed to dry out during this wash step Next the beads were combi ned with peptides dissolved in Coupling B uffer and incubated overnight at 4 C with end over end rotation to allow for conjugation Peptide coupled Sepharose beads were then washed with at least five gel volumes of Coupling B uffer, to remove unbound peptides. R emaining active groups were b locked by placing the beads in Blocking B uffe r (100 mM T ris HCl [ pH 8.0]) for 2 hours at room temperature. After the peptide coupled S epharose beads w ere blo cked, they were washed with three cycles of alternating pH solutions (five gel volumes per wash ) Each cycle consist ed of one wash with 100 mM acetic aci d/sodium acetate, pH 4.0, 500 m M NaCl, followed by a

PAGE 49

49 wash with 100 m M Tris H Cl (pH 8 .0 ) 500 m M NaCl The peptide coupled S epharose bea ds were then combined with 10 ml of anti sera, and incubate d for 4 hours at room temperature or overnight at 4 C with end over end rotation After antibodies were bound, the Sepharose beads were tra nsferred to a 5 ml plastic column and the flow through was saved for subsequent ELISA testing. The beads were washed three times with five gel volumes of P BS (phosphate buffered saline), and the OD 280 was monitored to determine when washes were sufficent The antibodies were e lute d with ice cold HCl (pH 2.5) solution Fractions of 900 l were collected into 1.5 ml Eppendorf tubes containing 100 l 1 M Tris HCl (pH8 .0) to neutral ize the antibody solutions immediately upon elution Sodium azide (0.02%) was added to prevent microbial contamination Each fract ion was then analyzed by ELISA (enzyme linked immunosorbent assay) ELISA p lat e wells were coated overnight at 4 C with pep tides from affinity purifi cation at a concentration of 100 ng/well dissolved in carbonate bicarbonate c oating buffer ( pH 9.6 ). ELISA plates were then blocked with 200 l of PBS containing 0.3% Tween 20 (PBS Tw) at 4 C until used for ELISA. T o determine the titer of purified antibodies each fr action was serially diluted by two fold starting at 1:500 to 1:32,000 in PBS Tw Pre immune and post immunization (final bleed) sera and the flow through from the affinity purification were included on each plate as controls. Whole cell lysates of S. mutans were then analyzed by Western blot using a dilution based on ELISA results of each fraction to det ermine the quality of the antibody and the success of the affinity purification. The SecY antibody resulted in a m ono specific reagent that recognized primarily a ~ 37 kDa band by SDS PAGE/Western blot A ~75 kDa band, which could be a SecY

PAGE 50

50 dimer, was occasionally detected as well as a ~20 kDa smear which has been reported by others to be breakdown products of SecY The YidC1 and YidC2 C terminal peptide anti sera were also affinity purified using the syntheti c peptides provided by Proteint ech coupled to CNBr activated sepharose Once affinity purified, the YidC1 antibody preparation recognized a singl e ~24 kDa band by Western blot. T he YidC2 antibody preparation recognized two bands after the first passage over the column, a ~28 kDa ba nd (desired band) and a ~ 20 kDa band which was also recognized by the pre immune sera. The 20 kDa band was only present in the firs t 12 column fractions and was not present in the flow through which still contained a considerable amount of antibody that recognized YidC2 T he refore, the flow through fraction was re applied to the column yielding a mono speci fic reagent that recognized a ~ 28 kDa band by Western blot The 20 resulting fractions were pooled and concentrated approximately seven fold using Ce ntiprep 1 0 columns. Creation of Strains SP13, SP14 and SP17: C Terminal Domain Swap s of YidC1 and YidC2. Splice overlap extension (135) was used to create a chimeric protein comprised of YidC1 (amino acids 1 229) and the C terminal tail of YidC2 (amino acids 247 to 310). Primers SP21F, with an Xba I site, and SP22RSOE (containing a 9 nucleotide overhang), were used to PCR amplify DNA encoding ( ) 131 to ( + ) 687 of yidC1 ( including the ribo some binding site) from S. mutans strain UA159 genomic DNA. This PCR product was then gel purified and referred to as fragment AB. DNA encoding the C terminal 63 amino acids of YidC2 was amplified by PCR from UA159 genomic DNA using p rimers SP22FSOE YidC 1 (with a 9 nt overhang corresponding to splice site with yidC1 ) and SP21R (with a Bsr GI si te) resulting in a fragment of y idC2 containing

PAGE 51

51 nucleotides (nt) 741 to 1,009, including 76 nt after the stop codon of the YidC2 open reading frame. This fragment w as also gel purified and referred to as fragment CD. Fragments AB and CD were combined through PCR overlap extension (135) using primers SP2 1F and SP21R. The product was initially cloned into pCR2.1 and then excised by enzyme digestion using Xba I and BsrG I. This fragment was then cloned into integration vector pBGE that had been digested with Xba 1 and Bsr G1. This vector was designed for chromosomal integration into the gtfA locus such that constructs are be expressed from the gtfA promoter and selected with a n erythromycin marker (133) facilitating the cloning of genes that may be lethal to E. coli Plasmid pBGE yidC1C2 Erm was used to transform AH378 (NG8 ::Kan R ) to generate strain S P13. A chimeric protein encompassing amino acids 1 247 of YidC2 and the C terminal amino acids 227 271 of YidC1 was constructed using splice overlap extension as described above for YidC1C2 The yidC2 gene fragment was amplified by PCR from UA159 genomic DNA using forward primer SP27F2, and re verse primer SP27RSOE (with a nucleotide overhang corresponding to yidC1 fragm ent), amplifying nucleotides ( ) 43 to (+) 741 of the yidC2 gene. This was gel purified and referred to as fragment AB. The yidC1 C te rminal fragment was generated by PCR from UA159 genomic DNA using forward primer SP28FSOE, and reverse primer SP28R (with a Bsr G I site) amplifying nucleotides (+) 685 to (+) 875 (this is 59 nucleotides downstream from the stop codon of yidC1 ). This PCR product was gel purified and referred to as CD. Fragments AB and CD were then used in a splice overlap extension PCR reacti on using primer s SP27F2 and SP28R. This PCR product was cloned into pCR2.1 and further pro cessed as with YidC1 C2 a bov e, except that instead of an Xba I site in primer SP27F2

PAGE 52

52 (a mistake was made during design ), the Xba I site in pCR2.1 was used. The resultant strain was named SP14 ( yidC2 ::Kan R gtfA:: yidC2 C1 Erm R ). Table 2 1 shows strain designations, and Table 2 2 list s primer sequences. Strain SP17, with yidC2 under the control of the gtfA promoter, was generated as the relevant positive control for experiments with SP13 and SP14. It was constructed in the same manner as SP13 and SP14 using primers SP27F2 and SP21R to PCR amplify yidC2 from UA159 chromosomal DNA. The yidC2 mutant strain that was used to create SP13, SP14, and SP17 was made from the S. mutans strain NG8. Location s of chimeric and control genes were verified by PCR using the forward primer SP37F, which binds in the promoter of gtfA and reverse primer AH31R for strains SP13 and SP17 or SP21R for strain SP14. Genes were also PCR amplified from chromosomal DNA and correct construction was confirme d by DNA sequencing. Conditional Expression of yidC2 In order to conditionally express yidC2 promoter fusion s were created between the promoters of the celA and celB genes and yidC2 See Figure 2 5 for a schematic diagram of the construction. T he region corresponding to the celA promoter (P celA 1 to 345 nt from celA start), including the ribosome binding site (RBS), was amplified by PCR from S. mutans strain UA159 genomic DNA using forward primer SP12F with an engineered Sma I site and reverse primer SP12R with an Nde I site which contains an ATG codon to facilitate promoter fusions ( see Table 2 2 for primer sequences) The celB promoter (P celB ) was amplified by PCR ( 1 to 263 nt from celB start) using primer s SP13F and SP13R with the same rationale as for the celA promoter. These primers were selected based on personal communication with Dr. Lin Zeng from the

PAGE 53

53 Burne Lab at the University of Florida who has experience with promoter fusions of P celA and P celB (133) The yidC2 gene was amplified by PCR from UA159 genomic DNA using forward primer SP14F ( with an engineered Nde I site) and reverse primer SP05 R ( with an engineered Sma 1 site ). Each PCR product was cloned into pCR2.1 (Invitrogen), and the resulting plasmids were restricted at a unique Bam HI site in pCR2.1 and the Nde I site ( engineered in to the prime rs ) to determine orientation The P celA and P celB clones containing ~ 45 bp fragment s were selected for large scale plasmid pre ps and restriction digestion The 45 bp fragment was removed by gel purification and t he larger fragment (~5.0 Kb) was reserved for a downstream reaction pCR2.1 clones containing yidC2 were selected that demonstrated a ~1.0 Kb fragment after Bam HI and Nde I digestion. This 1.0 Kb fragment was gel purified and ligated to the 5.0 Kb fragments containing either P celA or P celB from above, to create the promoter fusions P celA yidC2 and P celB yidC2 The constructs we re confirmed by DNA sequenc ing and then excised using the Sma I sites engineered in to primers, gel purified and cloned into Sma 1 digested pBGK2 ( same as pBGK in (136) but with t he Ap R marker removed by the Burne lab ) This is an integration vector containing a Kanamycin resistance marker and is designed for chromosomal integration into the gtfA locus of S mutans This resulted in plasmids pSP10 (P celA yidC2 ) and pSP11 (P celB yi dC2 ). The se plasmids were then used to transform the previously engineered mutant strain AH398 (31) resulting in an integration of the P celA and P celB yidC2 promoter fusions into the chromosome at the gtfA locus Transformants were selected on THYE agar plates containing kanamycin ( 5 00 g/ml ) The resulting strains were named SP10 (P celB yidC2 ) and SP11 ( P celA yidC2) The location s of the promoter

PAGE 54

54 gene fusion s in the chromosome were confirmed by PCR using ei ther forward primer SP16F ( binds within P celA ) or SP17F ( binds within P celB ) and reverse primer SP 16R ( binds after the stop codon of yidC2 ) Primers that were the r everse complements of SP16F and SP17F were designed ( SP16F RC and SP17F RC ) and used in PCR reaction s with SP18R which binds after the stop codon of gtfA to confirm that the promoter gene fusions were inserted in the opposite orientation as the gtfA gene. SP10 and SP11 were then used as the background strains to eliminate the endogenous copy of yidC1. SP20 and SP21 were created by natural transformation of SP10 and SP11 respectively with pCR2.1 yidC1 ::SpecR (see below) and selected on THYE agar plates containing 1000 g/ml spectinomycin. Deletion Construct ion of yidC1 using a Spectinomycin M arker. The yidC1 gene was eliminated by allelic replacement using a spectinomyci n marker To make the deletion construct s plice overlap extension (SOE) (135) was used to combine the upstre am and downstream fragments of yidC1 with a n intervening spectinomycin gene The upstream fragment was designe d not to interfere with the upstream gene rnpA which overlaps into the open reading frame of yidC1 The s pectinomycin marker was amplified by PCR from pDL278 (137) with primers SP25FSOE and SP25RSOE ( see Table 2 2 for primer sequence s ) resultin g in an 855 nt product The upstream region of y idC1 corresponding to 291 to +21 was ampli fied from UA159 genomic DNA using primers SP29F and SP24RSOE resulting in a 345 nt which bind s A PCR reaction was per formed with primers SP29F and SP25RSOE with the upstream yidC1 and spectinomycin fragments as the template through splice overlap extension (SOE) The downstream fragment of yidC1 was a mplified using primers SP26FSOE

PAGE 55

55 and AH25R with the forward primer binding 817 n t from the translational start site and directly after the stop codon of yidC1 The reverse primer bound 932 nt downstream fro m the forward primer This fragment was then comb ined with the previous SOE reaction using primers SP29F and AH25R to generate a 2,087 nt product, which was subsequently cloned into pCR2.1 (Invitrogen) to generate plasmid pCR2.1 yidC ::SpecR. Before transformation into S. mutans this plasmid was cut with Hind III resulting in a linear piece of DNA. T he yidC1 gene was replaced with the spectinomycin resistance gene through double crossover recombination. Growth Curves C ultures of S. mutans strain s SP22, SP17, AH378, AH374, SP13, SP14, AH412, and SP16 were inoculated from glycerol stocks into 10 ml THYE and grown overnight at 37 C (Table 2 1 for strain designations) Overnight cultures were diluted 1:20 into fresh THYE pH 7.0 broth without antibiotics and grown to an OD 600 of 0 .4. A 100 well Bioscreen C plate (Labsystems, Helsin ki, Finland) was filled with 180 l of pre warmed media (THYE pH 7.0 THYE pH 5.0 or THYE pH 7.0 with 3% NaCl ) Wells were inoculated in triplicate with 2 0 l of culture and covered with sterile minera l oil and pl aced in the Bioscreen C machine set to record optical density 600 nm every 15 minutes with shaking for 10 seconds before each reading, for 16 hours Doubl ing time (Td) was calculated by measuring the slope of the logarithmic growth phase using the formula Td = [ (t2 t1)ln (2)]/[ ln (OD2) ln (OD1) ] (31) Statistical analysis was performed using One Post Te st using the GrapPad Prism 4 program.

PAGE 56

56 Whole Cell Lysates Whole cell lysates were prepared from 10 ml cultures grown in THYE (Todd Hewett 3% Yeast Extract). Cells were pelleted at 3,210 x g for 10 minutes a nd washed once in 25 mM Tris HCl pH 7.5, and th en re suspended in 500 l of the same buffer. Each c ell suspension was then combined with 0.5 grams glass beads (0.1 mm BioSpec Products, Inc.) and shaken fo r 1 minute for two cycles in a min i bead beater ( Biospec Products ) and iced for 1 minute between cycles. Glass beads were allowed to settle for several minutes before the supernatant was transferred to a new tube. For SDS PAGE 50 l of each whole cell lysate was combined with 50 l 2X SDS sample buffer, boiled for 5 minutes and centrifuged at 13,0 00 rpm for 3 minutes in a Beckman tabl e top microcentrifuge Then 15 l was loaded on a 10% SDS polyacrylamide gel for Western blot analysis (described under Western blot protocol below). Cell Fractions and Gradient SDS PAGE Cells were harvested by centrif ugation at 3,210 x g for 10 minutes in a table top Beckman centrifuge from a 20 ml mid log (OD 600 0.5) culture grown in TDM (Terleckj defined media ) (138) containing 0.5% cellobiose or glucose. Proteins present in sterile filtered culture super natant ) were precipitated by combining 500 ed supernatant w trichloroacetic acid ( TCA ) and incubation on ice for 20 minute s. Precipitated proteins were pelleted in a tabletop microcentrifuge at 4 C at 13,000 rpm for 10 min ute s The p ellet s were wash ed twice with 300 l acetone and allowed to air dry in a 55 C oven for 10 minutes The dried pellet s were re suspend ed in 100 l 1 x SDS sample b uffe r. For other cellular fractions cell pellet s were harve sted and washed o nce in 10 ml B uffer A (10 mM Tris HCl [pH 6.8], 10 mM

PAGE 57

57 Mg acetate, in 25% sucrose or raffinose ) and then frozen at 20 C until further processing could be done. For the cell wall fraction, each frozen cell pellet was thawed and r e suspende d in 5 ml Buffer A 50 ul of 10 mg/ml lysozyme ) and 100 U/ml of mutanolysin ( ) Cells were incubated at 37 C for 45 min utes to 1.5 hours to protoplast the cells ( protoplast formation was monitored by Gram staining) P rotoplasts were pelleted by centrifugation at 3,210 x g for 10 min utes S u pernatant s from the cell wall digestion were f ilter with a 0.22 m syringe filter and TCA precipitate d ( 500 l with 500 l 20% TCA as described for culture supernatant s ) P rotoplasts were washed twice with 5 ml B uf fer A and re suspended in 1 ml Osmotic Lysis B uffer (50 mM Tris [pH7. 5] containing 10 mM MgSO 4 0.8M NaCl ) transfer red to a 5 ml po lycarbonate tube, and 40 ul of EDTA free protease inh ibitor cocktail (Complete Roche Tablet 25 x stock solution), 10 g/ml of DNase (10 l of DNase 1 mg/ml) and RNase (10 l of 1 mg/ml stock) were added Protoplasts were l yse d with two 10 s econd cyc l e s on setting 10 of a F isher Scientific Sonic D isme mbrator 100 (cells were cooled on ice between cycles ) The lysate s were transferred to a 1.5 ml E ppendorf tube and unlysed protoplast s and cell debris were removed by centrifugation in a tabletop micro centrifuge at top speed ( 13,000 rpm ) for 10 minutes The sup ernatant s were transferred to a 1.5 ml Beckman ultracentrifuge tube and centrifuge d at 100,000 X g (45K rpm) in SW50.1 rotor for 20 min utes at 4 C. The supernatant s, which represented the cytoplasmic fraction s were TCA precipitated as described for cultu re supernatant s and re suspend in 200 l 1X Native PAGE Sample B uffer (50 mM Bis Tris [pH 7.2], 50 mM NaCl 10% glycerol, 0.001% Ponceau S) The membrane pellet s were re suspended in 100 l of ice cold NativePAGE sample buffer using a Teflon homogenizer

PAGE 58

58 and a rounded Pasteur pipette to dislodge the sticky pellet s For SDS PAGE, each fraction was combined with 2 X SDS sample buffer, and boiled for 5 minutes followed by centrifugation at 13, 000 rpm for 3 minutes to remove insoluble material. Next each fraction was loaded on a BioRad Criterion 4 15% gradient SDS PAGE gel and electrophoresed for approximately 1 hour 25 minutes at 200 volts in a BioRad Criterion Cell connected to a BioRad PowerPac Basic Power Supply. Western Blotting After proteins w ere sepa rated by SDS PAGE they were transferred to an I mmobilon PDVF membrane (membrane was hydrated in 100% methanol according to the in a Hoefer Mighty Small TE 22 Mini Transfer Tank for 1 hour at 100 volts in Tr ansblot B uffer ( 25 mM Tris, 192 mM Glycine, 20% Methanol) After transfer membranes were blocked for 1 hour at room temperature or overnight at 4 C in PBS 0.3% Tween, 5% milk. Membranes were then reacted with primary antibody for 1 hour at room temperature Af finity purified YidC1 C terminal, YidC2 C termina l and YidC2 non C terminal antibodies were used at 1:4000, 1:8000 and 1:6000 dilutions respectively Affinity purified SecY peptide anti serum was used at a dilution of 1:1000 P eroxidase conjugated goat a ffinity purified anti rabbit IgG (Cappel MP Biomedicals, Solon Ohio) was used as the secondary antibody, which was reacted for 1 hour at room temperature at a 1:1000 dilution Western b lots were developed using the Amersham ECL kit from GE Healthcare. Res ults and Discussion Terminal Tail C ontributes to Stress Tolerance in S mutans A number of S. mutans strains were constructed in order to evaluate the function of yidC1 and yidC2 Whole cells lysates of each strain were e valuated by Western blot

PAGE 59

59 to confirm proper expression of YidC1 o r YidC2 (Fig. 2 6 ). The protein stain confirmed that similar levels of total protein were loaded for each strain. As can be seen from t he Western blot reacted with Y idC1 antiserum, all str ains recognized a ~ 24 kDa band except for those strains in which yidC1 had been replaced with an antibiotic marker, yidC1 ::Erm, SP15 ( yidC1::Spec ) and SP16 ( yidC2 C::Erm yidC1::Spec ). Also strain SP14, containing the chimeric protein Y id C2 C1 resulte d in two bands of reactivity one at the expected ~ 24 kDa corresponding to wild type YidC1 an d one slightly larger band at ~ 25 kDa corresp onding to the chimeric YidC2C1 protein The Western blot reacted with the Y idC2 C term inal antibody showed the expected reactivity with all strains containing the YidC2 C terminal tail reacting with a band of the correct size Wild type full len gth YidC2 migrates as a ~28 kDa band and was observed for NG8, yidC1 SP15, and SP17 (Table 2 1 for strain designation s). Strain SP13, containing the chi meric YidC1 C2 protein, demonstrated a slightly smaller band of ~27 kDa. Reactivity with Y idC2 TM 2/3, a YidC2 peptide antibody that recognizes a non C terminal epitope, showed reactivity with a ll strains as would be e xpected, recognizing a ~ 28 kDa band in NG8, yidC1 SP15, and SP17. Strain s yidC2 C (AH412) and SP16 ( yidC2 C, yidC1 ) did not react with the Y idC2 C term inal antibody but demonstrated a ~22 kDa band reactive with the non C terminal Y idC2 antibody. Once proper YidC expression was confirmed, t hese strains were evaluated for their abili ty to tolerate acid and osmotic stress, by growth in THYE pH 5.0, and THYE 3% NaCl, compared to growth under non stress conditions in THYE pH 7.0. Me an doubling times are shown in T able 2 3. Results indicate that in strain SP13 ( yidC2

PAGE 60

60 gtfA :: yidC1 C2 ) chimeric YidC2C1 was able to significantly restore acid tolerance to the yidC2 strain as well as c omplete unaltered YidC2 (SP17, yidC2 gtfA:: y idC 2 ), while salt tolerance was restored to some extent it was not to a significant level. In contrast, the growth defect of strain SP14 ( yidC2 gtfA :: y idC2 C 1 ) was significantly worse than that observed upon complete deletion of yidC2 ( strain AH378 ) under no n stress conditions. This indicates that replacing the C terminal tail of Y idC2 with that of Y idC1 destroys its ability to function normally perhaps by affecting necessary protein protein interactions The functional relevance of the YidC2 C terminus is further supported by the fact that placing it onto YidC1 confers on YidC1 the ability to restore acid tolerance to the yidC2 mutant. Of note SP17 did not grow like the wildtype strain under stress conditions when YidC2 was expressed from the gtfA pro moter This suggests that the yidC 2 promoter probably also contributes to stress tolerance, perhaps by allowing for increased expression of yidC2 under certain conditions as was demonstrated in studies of the Lia SR TCS (45) The yidC2 C terminal deletion strain, AH412 showed similar acid and salt sensitiv ity as strain AH378 with a complete deletion of yidC2 H owever AH 378 grew more slowly tha n the AH412 under non stress conditions. This suggests that C terminal tail has a function that confers stress tolerance and that in the absence of stress additional non C terminal functions of YidC2 contribute to normal growth Deletion of yidC1 in the AH412 background (strain SP16) further exacerbated the growth defects of the YidC2 C terminal deletion. This suggests that in the abs ence of th e C terminal tail of YidC2, Y idC1 can play a compensatory role in acid and salt t olerance. P erhaps the overall amount of YidC proteins is important. Attempts to delete yidC1 from strains SP13 and SP14 were unsuccessful. This was somewhat

PAGE 61

61 surprising in t he case of SP13, in which YidC1C2 was able restore stress tolerance to the yidC2 mutant strain. Perhaps, the level of expression was an issue and use of the yidC2 promoter to express Y i dC1C2 would have resulted in full complementation. Alternatively, Yid C1 and YidC2 may cooperate in a balanced manner that cannot be replicated by YidC1C2 in the absence of endogenous Y idC1 and Y idC2. In addition to growth in broth s t rains expressing chimeric YidC1 C2 (SP13) an d YidC2 C1 (SP14) were evaluated for their ability to restore growth of the yidC2 mutant on pH 5.0 THYE agar plates (Figure 2 7). For comparison, strains SP22 (NG8, gtfA:: Erm ) SP17 ( yidC2 gtfA::yidC2 ) and AH412 ( yidC2 C ) were included Strain SP22, which yidC2 is expressed from the end ogenous promoter showed the best growth. When the yidC2 mutant was complemented with yidC2 in the gtfA locus and expressed from the gtfA promoter (SP17), growth was restored but not to wildtype levels. Deletion of the C terminal tail of YidC2 ( with the deletion in the chromosomal yidC2 gene) had an intermediate phenotype on agar. In contrast, growth in broth culture of this strain was similar to that of the full yidC2 mutant This suggests that t he C terminal tail of YidC2 is less necessary fo r acid tolerance when the bacteria are grown on solid medi a perhaps due to differences in oxygen tension When Y idC1C2 was expressed in the yidC2 background (SP13), it was able to restore growth of the yidC2 mutant as well as wild type yidC2 (SP17 ). Ho wever, w hen the C terminal tail of YidC1 was place d o nto YidC2 (SP14) no complementation of growth of the yidC2 mutant was observed

PAGE 62

62 Conditional Expression of yidC2 In order to better understand the functions of redundant pathways for protein secretion and membrane biogenesis in S mutans a conditional e xp ression system was needed, to allow potentially lethal combinations of deletions to be made. A conditional expression sy stem was developed for yidC2 using the promoters from the cellobiose operon, PcelA and PcelB It was shown by Zeng et. al. 2009 using promoter fusions of PcelA and PcelB with the CAT reporter gene ( c hloramphenico l a c ety l t ransferase) followed by CAT assay s after growth in different sugars that expression from these promoters is control led by carbon catabolite repression (CCR) in S. mutans strain UA159 with almost no expression in the presence of the repressing sugars glucose, fructose and mannose (133) In the current study, the constructs made were in the S. mutans NG8 backgroun d. W hile similar to UA159, there are some differences in CCR between the stains (personal communication, L. Zeng and R. A. Burne). Strains SP10 an d SP11 were created first, with the yidC2 gene fused to PcelB or PcelA respectively, and then placed into th e gtfA locus of the yidC2 mutant ( AH398 ) chromosome (Table 2 1). The yidC1 gene was then replaced with a spectinomycin marker in SP10 and SP11, to create SP20 and SP21 re spectively To determine which carbon source has the most repressing effect on yidC2 expression when under the control of PcelA (SP11/SP21) or PcelB (SP10/SP20) d ifferent growth conditions were tested followed by Western blot with YidC2 C term inal and YidC1 C term inal antibodies. T he Western blot results of SP10 and SP20 grown in TDM with 0.5% ce llobiose (inducing conditions) compared to growth in TDM 0.5% glucose (not shown), 0.5% fr uctose, 0.5% mannos e, and THYE are shown in Figure 2 8. O vernight growth in TDM 0.5% mannose had the

PAGE 63

63 most repressing effect on yidC2 expression w hile growth in TDM 0.5% fructose had only a minor repressing effect Shown in Figure 2 9 are the Western blot res ults for strains SP11 and SP21 gr own under the same conditions mentioned above. Somewhat surprisingly, PcelA expression of yidC2 in the absen ce of yidC1 (SP21) was not repressed by overnight growth in mannose to the same extent as PcelB controlled expression of yidC2 (SP20). However, when endogenous yidC1 was present in strain SP11, expression of yidC2 under the control of PcelA was more repressible (Figure 2 9, last two lanes). This was also true for SP20 and SP10. SP10 demonstrated repression of yidC2 in the presence of glucose In contrast no repression of yidC2 by glucose was observed for strain SP20 (not shown) and repression by mannose occurred only a fter overnight growth. The promoter fusions made in these stu dies were in an NG8 background. In contrast to what Zeng found in UA159, expression from the celA promoter in the NG8 background was very leaky in glucose, fructose and m annose, while expression from the celB promoter was more tightly regulated, but only in TDM 0.5% mannose (See Western blots in Figures 2 8 and 2 9). Various c ell fractions were prepared from strains NG8, yidC1 yidC2 and SP10 and compared to strain SP20 in which yidC2 expression is limited in a yidC1 negative background. Figure 2 10 shows the Western blot analysis of the membrane fra ctions prepared from cells grown in TDM 0.5% mannose or 0.5% cellobiose reacted with YidC2 C term inal YidC2 TM 2/3, and YidC1 C term inal antibodies. Membranes from cells grown in TDM 0.5% mannose, where yidC2 expression is low (SP20) and TDM 0.5% cellobiose, where yidC2 expression is high are shown for SP10 and SP20 at different time points. R epression of yidC2 was observed in SP10 (in which yidC1 is

PAGE 64

64 present) sooner and to a gr eater extent than in SP20, in which yidC1 is absent and yidC2 expression is regulated by CCR through the PcelB promoter. Th e Western blots also confirm ed that the amount of Y idC2 was negligible in SP20 when cells were grown in TDM 0.5% mannose for 7 hours. C ell wall membrane and cytoplasmic fractions were analyzed by one dimensional gradient SDS PAGE followed by silver stain. T he protein profile s of cell wall fractions from the respective strains grown to late log phase in TDM 0.5% mannose are shown in Figure 2 11 For strain s SP10 and SP20 cell s were grown in TDM 0.5% mannose for 5 ho urs or 7 hours respectively. T he protein profiles of cytoplasmic and membrane fractions from the various strains grown in TDM 0.5% mannose or 0.5% cellobiose where indicated are shown in Figures 2 12 and 2 13 Only minor difference s as reflected by 1D SDS PAGE analysis, were observed in the cell wall, cytoplasmic or membrane fractions between the wildtype strain and the yidC1 or yidC2 mutants or between the single mutants and SP20, in which Y idC2 is depleted in a yidC1 mutant background Since YidC proteins may be involved in assembly of multimeric protein complexes, these strains were assessed further using the Blue Native PAGE technique described in Chapter 4. In a recent proteomics study of Y idC depletion in E. coli under aerobic compared to anaerobic conditions by Price (139) the effects were less pronounced under anaerobic growth conditions Since S. m utans is a facultative anaerobe, effects of Y idC depletion in this bacterium could be influenced by oxygen as well and more readily observed at the functio nal level of individual enzymes The complete mechanisms of CCR are not known is S. mutans however it is clear from work by the Burne Lab at the University of Flo rida that it is different from CCR in E.

PAGE 65

65 coli and involves EIIAB Man a membrane associate d protein of the mannose/glucose PTS system (140) Mutation of EIIAB Man results in loss of CCR, increased cellobiose PTS activity, decreased fructose PTS activity, decrease d biofilm formation, and loss of genetic competence (141) The S. mutans yidC2 mutant has a pronounced growth defect, with decreased growth rate and cell yield in defined media containing a number of different sugars (this effect is less pronounced when cells are grown in complex media THYE), decreased biofilm formation, and a loss o f genetic competence. It would be interesting in future studies to evaluate the effects of yidC1 or yidC2 on CCR which likely involve s a number of membrane components This could be tested by deletion of yidC1 or yidC2 from strains with CAT fusions to P celA or PcelB and assay of CAT activity after growth in different repressing sugars Use of the P celB or P celA promoter s is a promising tool for conditional expression of essential genes in S. mutans These could be optimized further though additional mutation for better control over gene expression in pathways tha t influence CCR, or by using different combinatio ns of repressing sugars, or adjusting the concentrations of repressing sugar s It remains to be seen if Y idC1 or Y id C2 are involved in CCR. The results of these studies suggest this might be the case. T herefore PcelA and PcelB may be more effective at enabling conditional expression of genes that do not contribute to CCR It i s also possible that the effects of carbo n source seen in SP10 and SP20 are related to a stress response that occurs in S. mutans and that affects CCR when essential genes are eliminated. In that case use of CCR controlled promoters for conditional expression of essenti al genes may be more comp licated and difficult to

PAGE 66

66 achieve. Additionally, it i s possible the discrepancies seen with repressing sugars are related to differences in CCR between UA159 and NG8. Figure 2 1 Clustal W sequence alignment of S. mutans YidC1 and YidC2. Sequences predicted to be transmembrane (TM) domains by TMPred are boxed and labeled 1 6 YidC1 and YidC2 are predicted to be lipoproteins that are process ed by SPaseII at amino acid 23 in YidC2 and amino acid 20 in YidC1 (indicated by an arrow). N terminal sequencing showed that YidC1 and YidC2 are f urther processed resulting in 218 a.a. and 253 a.a. proteins respectively (142) There is approximately 30% identity and 75% similarity between YidC1 and YidC2 Differences are seen between the C terminal tails, with YidC2 containing a longer more basic tail (+17) tha n YidC1 (+10).

PAGE 67

67 Figure 2 2. Schematic representation s of yidC1 yidC2 and yidC2 and construc tion of chimeric proteins are shown. YidC1C2 was produced by strain SP13 and YidC2 C1 was produced by strain SP14. Transmembrane domains are indicated in white. Predicted s izes in kilodaltons o f mature proteins are indicated with the c orresp onding nucleotide seq uences and number s of amino acids, listed below each construct in parentheses.

PAGE 68

68 Figure 2 3 Membrane topology model of YidC1 with location of C terminal peptide used to make C terminal antiserum Transmembrane domains are modeled based on the TMPred p rediction program A synthetic peptide corresponding to the a mino acids located in bold circles was used to prod uce rabbit C terminal peptide antiserum against YidC1 Amino acids are color coded in grey scale based on chemical properties as indicated by t he key at the top of the figure Predicted SPaseII cleavage site is indicated by an arrow

PAGE 69

69 Figure 2 4 Membrane topology model and location of peptides used to make antisera against YidC2 The TMpred prediction program was used to predict the loc ation of transmembrane domains The p redicted SPaseII cleavage site is indicated by an arrow Residues in bold circles correspond to peptides used to immunize r abbits to produce peptide antibodies Of t he two non C terminal peptides, only the one corres ponding to cytoplasmic loop between TM 2 and 3 ( Y idC2 TM 2/3) was used in later experiments The antibody that was produced using the extracellular peptide between TM domains 3 an d 4 had only weak reactivity against the S mutans YidC2 protein The Yid C2 C terminal peptide use d to produce C terminal antiserum is a lso indicated by bold circles Amino acids are color coded in grey scale based on chemical properties See key at top of figure for code.

PAGE 70

70 Table 2 1 Bacteria l strains and plasmids used in this study Strain N ame Relevant C haracteristic( s) Source Streptococcus mutans NG8 Wild type Brady lab AH398 yidC2 :: e r m (31) AH378 yidC2 :: kan A. Hasona AH374 NG8 yidC1 :: erm (31) AH412 NG8 yidC 2 255 310 :: erm (123) SP01 AH398/pDL289 ( Kan R ): yidC2 1 50 E. coli yidC 59 548 (142) SP10 AH398 gtfA ::P celB yid C2 kan This study SP11 AH398:: gtfA:: P celA yid C2 kan This study SP12 NG8 gtfA : : P gtfA :: yid C 1 1 249 yid C2 246 310 :: Erm This study SP13 AH378 gtfA :P gtfA :: yid C 1 1 2 49 yid C2 246 310 :: Erm This study SP14 AH378 gtfA :P gtfA :: yid C2 1 2 47 yid C1 228 271 :: Erm This study SP15 NG8 yidC1 :: sp c This study SP16 yidC 1 :: sp c This study SP17 AH378 gtfA : : P gtfA yidC2 : : erm This study SP20 yidC1 : : sp c This study SP21 yidC1 :: sp c This study SP22 NG8 gtfA :: erm This study SP23 SP13 yidC1 :: s p c* This study SP24 SP14 yidC1 :: sp c* This study SP25 SP01 yidC1 :: sp c* This study Plasmids pDL289 E. coli Streptococcus shuttle vector; Kan r (143) pDL278 E. coli Streptococcus shuttle vector, Spc r ( aad9 ) (137) pBGK2 Streptococcus integration vector gtf A locus (same as pBGK, amp R was deleted) (136) pBGE Streptococcus integration vector gtf A locus, genes are expressed from the gtfA promoter in the chromosome (133) pSP10 pBGK2 with P celA yidC2 cloned into Sma 1 site This study pSP11 pBGK2 with P celB yidC2 cloned into Sma 1 site This study pCR2.1 T/A cloning vector, multiple cloning site, Amp r Kan r Invitrogen pCR2.1 yidC1SpecR pCR2.1 with yidC1::spc for allelic replacement of yidC1 This study Antibiotics are abbreviated as; erythromycin Erm kanamycin Kan, ampicillin Amp spectinomycin Spc. *Attempts to delete yidC1 in this background failed or mutants were barely viable. In the case for SP25, mutants took 4 days to grow.

PAGE 71

71 Table 2 2. Oligonucleotides used in this study Oligonucleotid e primers SP12F GTCATAGTCAATCCCGGGTTATTTTGATAAAT SP12R TTAGACATATGTCTACCTCCTTTTTCTTA SP13F CTTCAAATGCCCGGGACATTTAAGATTAAT SP13R GTTTTGCCATATGATTCTAATCTCCTTT SP14F CAAAGGAAGATTCATATGAAAAAAATTTA SP05R GATTAGGTCCCGGGGTTGATAACAGCTT SP16F GATGCCAATTCTAGCTTTTATA SP17F CAGCAAATGATAAGTCAAAAATA SP16F RC TATAAAAGCTAGAATTGGCATC SP17F RC TATTTTTGACTTATCATTTGCTG SP18R TTAGTTCAACCATAGTCTCTC SP21F CCTTATTTAGGCACTCTAGATTTT SP21R GCTTCCGTTTGTACATGAGACGAT SP22FSOE YidC1 TATCAGGTCACAAACCATATCATTAAACCAAAA SP22RSOE YidC2 ATGGTTTGTGACCTGATAAGCATTTGAGAC SP27F2* ACGAGATCTCAGTAAAATAGATAGGTATTATTAA SP27F2 2 ACGATCTAGACAGTAAAATAGATAGGTATTATTAA SP27RSOE CTGGAAGACGATCAATTGTTGGATGAGACCAAA SP28FSOE CAATTGATCGTCTTCCAGATTCTGTTGTTAAATAA SP28R CGCTGTACACTTTCAATAGCTTCTTCAAC AGTTTT SP29F CTGAAGGACGAAGTGTTGCCAAT SP24RSOE CATAGTTGTAAGTGCTGCAACAGCTAATCCAAT SP25FSOE GCAGCACTTACAACTATGGATATAAAATAGGTA SP25RSOE TTGTTCTCCGTTTCCACCATTTTTTCAATTTTT SP26FSOE GGTGGAAACGGAGAACAAATCATGGTATTATT AH25R CTTTCAACTCCTTCAATGCCAGTAAT AH31R AGCTTATTGC TTATGGTGACGC SP36F CGCCAGCCGCAATAAGAACGATT SP37F GTGGCTGTTATTTTTAGGTTGAA a Underlined sequences correspond to engineered restriction enzyme sites (6 nucleotides) or overhangs for splice overlap extension (9 nucleotides) *Primer SP27F2 was intended to contain an Xba I site, SP27F2 2 was designed to replace SP27F2 and contained an Xba I site.

PAGE 72

72 Figure 2 5 Schematic diagram of the construction of promoter fusions of P c elA and P celB with yidC2 (see text in Materials and Methods for exact protocol) Plasmid pBGK (136) is shown, however the plasmid that was used (pBGK2) had th e Ap marker removed for use in s treptococci This plasmid is an integration vector, which places the cloned gene into the gtfA locus in the S. mutans chrom osome through double crossover recombination The promoter fusions were cloned in the opposite orientation as the gtfA gene so there would be no read through from upstream promoters The o rientation was confirmed by PCR using primers SP16F RC (P celA ) or SP17F RC (P celB ) and SP18R which binds after the stop codon of gtfA

PAGE 73

73 Figure 2 6 Western blot of whole cell lysates of indicated strains reacted with antibodies against YidC1 and Yid C2 Top panel shows an Aurodye TM protein stain used as a loading control Bottom panels show Western blot results with antisera of the indicated specificity. Strain genotypes are as follows: SP13 yidC2 gtfA::yidC1C2 SP14 yidC2 gtfA::yidC2 C 1 SP15 yidC1 ::Spc R SP16 yidC2 C, yidC1 SP17 yidC2 gtfA::yidC2. See Table 2 1 for detailed information.

PAGE 74

74 Table 2 3. Mean doubling times (mins.) a of S mutans yidC mutants and complemented strains under non, acid and osmotic stress conditions. Strain Genotype THYE pH 7.0 THYE pH 5.0 THYE 3% NaCl SP22 NG8 gtfA::erm 73 4.7 167 6.6 240 17.6 SP17 yidC2 gtfA::yidC2 77 1.5 285 4.2 334 27.5 AH378 NG8 :: kan 114 1.7 346 19.4 401 24.0 AH374 NG8 yidC1::erm 77 11.9 186 5.6 266 27.8 SP13 AH378 gtfA::yidC1C2 80 1.5 282 19.0 346 2.7 *** SP14 AH378 gtfA::yidC2C1 171 5.3 390 41.7* NG AH412 NG8 yidC2 C ::erm 83 10.0 367 0.0* 399 40.0 SP16 AH412 yidC1::spc 143 3.2 431 17.8* NG a Doubling times were calculated based on growth curves completed in triplicate in a Bioscreen C 100 well micro titer plate and OD600 monitored in a Bioscreen C machine. NG indicates no growth. Statistical differences by One way ANOVA are indicated comp ared to; SP22 = P< 0.001, *= P< 0.01, *** = P < 0.05; AH378 = P<0.001, = P< 0.01, =P < 0.05; AH412 =P < 0.05.

PAGE 75

75 Figure 2 7 Growth on THYE pH 5.0 plates of wild type, yidC2 yidC2 C S. mutans and complemented mutant strains Panel on left shows growth after 48 hour incubation at 37 C in a 5% CO 2 incubator. Panel on right provides a key for location of strains.

PAGE 76

76 Figure 2 8 Western blot s of whole cell lysates from Y idC2 depletion strains SP20 and SP10 grown in TDM 0.5% sugar or in THYE SP10 and SP20 contain promoter fusions of yidC2 to PcelB SP20 was made fro m SP10 by replacing yidC1 with a spectinomycin marker NG8, YidC1 C term control) and YidC2 C term control) were used as negative controls f or antibody reactivity Western blots w ere reacted with the indicated antibody and developed by ECL O/N is an abbreviation for overnight culture. Figure 2 9 Western blot s of while cell lysates from Y idC2 depletion strains SP21 and SP11 grown in TDM 0.5% sugar or THYE SP11 and SP21 contain promoter fusion s of yidC2 to P celA SP21 was made from SP11 by replacing yidC1 with a spectinomycin marker NG8, yidC1 ( YidC1 C term control) and yidC2 ( YidC2 C term control) were used as negative control s for antibody reactivity Western blots w ere reacted with the indicated antibody and developed by ECL.

PAGE 77

77 Figure 2 10 Western blot s of membrane fractions from YidC2 depletion strains SP10 and S P20 grown in TDM 0.5% sugar Cells were g rown to mid log phase in the indicated sugar or in the case of SP10 and SP20 grown in mannose for the indicated time s Proteins were separated on 12% SDS PAGE gels and reacted with the indica ted anti sera When YidC2 expressi on was high (0.5% cellobiose), an unprocessed 34 kDa band was see n as well as a number of lower molecular weight breakdown products.

PAGE 78

78 Figure 2 11 Cell wall extracts of S mutans wildtype yidC mutant strains or Y idC2 de pletion strains SP10 and SP20 C ells were grown in TDM 0.5% mannose Extracts were prepared by digestion of the cell wall with mutanolysin and lysozyme Proteins were precipitated with TCA and separated on a 4 15% gradient Criterion gel and stained using the Silver Stains Plus Kit (BioRad).

PAGE 79

79 Figure 2 12 Cytoplasmic fraction s of S mutans wild type yidC mutant strains or Y idC2 depletion strains SP10 and SP20, grown in TDM 0.5% of the indicated sugar Proteins were precipitated with TCA and separated on a 4 15% gradient Criterion gel and stained using the Silver Stain P lus kit from BioRad.

PAGE 80

80 Figure 2 13 Membrane fractions of S mutans wild type, yidC mutant s, or Y idC2 depletion strains SP10 and SP20 Cells were grown in TDM with 0.5% of the indicated sugar Proteins were separated on a 4 15% g radient Criterion gel and stained using the Silver Stains Plus kit from BioRad.

PAGE 81

81 CHAPTER 3 F 1 F O ATP SYNTHASE Rationale for Study In S. mutans the F 1 F o ATPase plays a large role in acid tolerance by pumpin g protons out of the cytoplasm. This contributes to maintenance of a hospitable pH for glycolytic enzymes during acid stress conditions created by metabolic end products. The F 1 F o ATPase is also important for maintenance of the proton motive force (PMF) Assembly of the F 1 F o ATP synthase has been studied in E. col i and insertion of the integral membrane components were shown to involve the SecYEG YidC, SRP and YidC only pathwa ys (29 30, 144 145) The mechanisms of insertion of the F 1 F o ATPase in S. mutans are not known; however mutations of ffh or yidC2 decrease membrane associated ATPase activity and acid tolerance (31, 74) In previous work a deletion of yidC1 had no apparent effect on mem brane associated ATPase activity or acid tolerance (31) leading to the hypothesis that in S. mutans YidC2 and the SRP pathway are involved in the assembly of the F ATPase, while YidC1 is not In this study, the ability of YidC1 and YidC2 to insert components of the F 1 F o ATPase was evaluated in the E. coli YidC depletion strain JS7131 in collaboration with Ross Dalbey at Ohio State University (142) A sequence alignment of the YidC proteins of S. mutans and E. coli usi ng Clustal W show s that the 5 C terminal transm embrane domains are conserved (F igure 3 1 ). Genes were engineered for expression of c himeric p roteins of E. coli Y idC and YidC1 and YidC2 from S. mutans The m embrane topologies of the YidC proteins are sho wn in Figure 3 2 A. E. coli YidC demonstrates six tra nsmembrane (TM) domains and a large periplasmic loop between TM 1 and 2 This architecture is conserved in gram negative bacteria YidC1 and YidC2 from S.

PAGE 82

82 mutans are predicted to be lipoproteins that are processed by SPaseII resulting in mature proteins with five transmembra ne domains ( Figure 3 2 A ) To ensure proper membrane targeting in heterologous host s chimeric proteins between the E. coli YidC and YidC1 and YidC2 from S. mutans were c onstruc ted as indicated in Figure 3 2 B The chimeric S. mutans YidC1 and YidC2 proteins were then evaluated for their ability to complement growth in the E. coli YidC depletion strain JS7131 and for their ability to function in the insertion and assembly of F 1 F o ATP synthase subunits in E. col i. This was measured by ATP hydrolysis and proton motive force assays at the University of Florida and by protease accessibility assays with model YidC subst rates at Ohio State University. Additionally, a chimeric E. coli YidC protein was evaluate d for its ability to restore stress tolerance to a yidC2 mutant in S. mutans The contribution of YidC1 and YidC2 to ATPase activity in membranes and whole cells from S. mutans was also evaluated u sing inverted membrane vesicles a nd permeablized whole cells The membranes and permeablized cells were prepared from the strains descri bed in Chapter 2. These produced chimeric proteins YidC1C2 (SP13) an d YidC2C1 (SP14), as well as YidC2 lacking the C term inus (AH412). Materials and Methods Chimeric S. mutans y idC2 y idC E. coli Construction All plasmid preps and gel extractions were completed using QIAGEN QIAquick kits. Standard molecular cloning was performed using enzymes and buffers from New England Biolabs DNA corresponding to amino acids 60 548 of E. coli Yi dC was amplified by PCR using plasmid DNA from pACYC184 YidC ( provided by Ross Dalbey Ohio State University ) with vent DNA polymerase and AAACTG CCTAGG an engineered Avr II site indicated,

PAGE 83

83 CCCGGG a n engineered Sma I site indicated. The resultant 1.4 Kb product was cloned into the Zero Blunt TOPO PCR cloning vector from In vitrogen and transf ormed into I nvi trogen chemically competent TOP 10 E. coli cells. Transformants were screen ed for orientation by enzyme digestion. A clone with the correct insert orientation was subjected to enzyme digestion using the Hind III site in the vector and the engineered Avr II site in t he PCR product to release a 1.5 Kb fragment Next DNA corresponding to t he promoter region and encoding the first 50 amino acids (a.a.) of Y idC2 was amplified by PCR using Vent polymerase from S. mutans NG8 geno mic DNA. The CCCGGG AAAT engineered with a Sma I site (underlined) AAAA CCTAGG GATA ACACTTCC CAT Avr II site. The resulting PCR produc t was ligated to pACYC184 that h ad been restricted with Eco RV through a blunt ligation reaction. The ligation product was transformed into Invitrogen chemically competent TOP 10 E. coli cells. T ransformants were screened for insert orien tation by enzyme digestion A clone with the correct orientation was restricted with Avr II and Hind III followed by g el purification to re move the resulting 80 bp fragment The 1.5 Kb Hi n d III Avr II fragment containing the DNA of E. coli YidC was then ligated to the gel purified DNA containing pACYC184 and part of the yidC2 gene The chimeric yidC2 1 50 yidC EC 59 548 (2.2 Kb) gene, encoding the protein 50YidC, was then excised from pACYC184 by Sma I digestion and cloned i nto Sma I digested pDL289 ( E. coli to Strept ococcus shuttle vector ) resulting in plasmid pSP4. The c onstruct ion was

PAGE 84

84 confirmed by DNA sequencing and pSP4 was transformed into AH398 ( S. mutans yidC2 mutant) by electropo ration as in (146) to create strain SP01. E. coli YidC Depletion Strain JS7131 with pACYC184 S. mutans y idC 1 and y idC2 Constructs. Construction of pCR2.1 yidC1 and pCR2.1 yidC2 These plasmids were made by Dr. Adnan Hasona in the Brady lab (University of F lorida) for use by Ross Dalbey in the generation of S. mutans E. coli chimeric proteins at Ohio State University. The TOPO TA Cloning kit from Invitrogen was used to clone S. mutans yidC1 and yidC2 into pCR 2.1 TOPO YidC1 and yidC2 were amplified by PCR from UA159 genomic DNA using the following primers for yidC1 yidC2 TTATG The PCR pr oducts were cleaned usi ng QIAGEN QIAquick kits and cloned into pCR 2.1 TOPO following the C himeric E. coli S. mutans y idC 1 and yidC2 constructions Genes encoding chimeric proteins between E. coli yidC and S. mutans yidC1 and yidC2 were constructed at Ohio State University, in the Ross Dalbey lab. This information was adapted from (142) The pCR 2.1 vector has a Not I (compatible with EagI ) restriction enzyme site shortly upstream of cloned yidC1 or yidC2 and an Eag I site shortly downstream of them. To construct pACYC184 247YidC1, the Eag I fr agment from pCR2.1 y idC1 was inserted into pACYC184 YidC downstream of the yidC stop codon, to yield plasmid pACYC184 Yi dC YidC1. Additional DNA sequences encoding amino acids after the 247th of YidC and before the 26th of YidC1 were deleted by oligonucleotide directed loop out mutagenesis using the Quick Change method. The

PAGE 85

85 same strategy was used to create pACYC184 YidC1 eliminating the DNA sequences between the 1 st amino acid of E. coli YidC and the 2nd amino acid of YidC1 in pACYC184 YidC YidC1 A similar strategy was used to construct pACYC184 247YidC2 and pACYC184 YidC2. To construct pACYC184 YidC1 YidC2, a n Eag I s ite was introduced into pACYC184 YidC2 upstream of its yidC promoter region (n ote that the previous Eag I site was removed during construction of pACYC184 YidC2). Then the Eag I fragment from pACYC184 YidC2 containing the yidC2 gene was inserted into pACYC1 84 YidC1 downstream of yidC1 Therefore expression of 247YidC1, 247YidC2, YidC1 and YidC2 were all under control of the E. coli yidC promoter. Recombinant plasmids were transformed into the E. coli YidC depletion strain JS7131 (147) Growth Curves S. mutans Overnight cultures of S. mutans wild ty pe strain NG8 AH374 ( yidC1 ) AH398 ( yidC2 ), and SP01 ( AH398 with pSP4 encoding 50YidC ) were diluted 1:20 in THYE pH 7.0 without antibiotics and grown to an OD 600 of 0.4 A 100 well Bioscreen C plate (Labsystems, Helsinki, Finland) was filled with 300 of pre warmed media (THYE pH 7.0 T HYE pH 5.0 or THYE pH 7.0 with 4% NaCl ) Wells were inoculated in triplicate with 30 of culture and grown for 16 hours at 37 C with absorbance at 600 nm recorded every 15 minutes. E coli YidC Depletion Strain JS7131 To determine whether S. mutans YidC1 or YidC2 could restore growth to the E. coli YidC depletion strain JS7131 in broth cells were grown overnight at 37 C in LB medium containing 0.2% arabinose. Overnight cultures were diluted 1:10 int o fresh LB medium containing 0. 2% arabinose and g rown

PAGE 86

86 to an OD 600 of 0.7 0.8. Cultures were washed once with LB (no sugar) and suspend ed at a 1:20 dilution in LB containing 0.2% glucose Cultures were grown at 37 C for 2 hours in order to deplete YidC. After YidC depletion, cultures were diluted 1:25 into fresh LB with 0.2% glucose (repressing sugar) or LB with 0.2% arabinose (inducing sugar) applied in triplicate to a 100 well Bio screen C plate which was inserted into a Bioscreen C machine (Labsystems, Helsinki, Fi nland), set at 37 C to read OD 600 every 15 min utes for 5 h ours with shaking for 10 min utes in between readings. S pectinomycin ( ) and chloramphenicol ( ) were used where appropriate. Growth on LB Agar Plates of E. coli JS7131 E xpressing S. mutans YidC P roteins C omplementation of growth of JS7131 by S. mutans YidC1 and YidC2 and derivatives on LB agar plates was also tested The YidC depletion strain JS7131 and derivatives were grown at 37 C overnight in LB medium sup plemented with 0.2% ara binose After being washed twice with plain LB the overnight culture s were streaked onto LB agar plates containing 0.2% arabinose or 0.2% glucose and incubated overnight at 37 C. Preparation of Inverted Membrane Vesicles E coli Strain JS7131 harboring S. mutans yidC1 or y idC2 constructs was grown in LB medium containing 0.2% arabinose. E. coli LB medium containing 0.2% glucose A fter overnight incubation at 37 C, bacterial cells were pelleted by centrifugation at 2,500 x g and wa shed once in LB medium. Cells were re suspended in plain LB medium and transferred to 1 liter of LB medium with

PAGE 87

87 spectinomycin where appropriate Cultures were grown at 37 o C until they reached an OD 600 of 0.5 0.55 (2.5 5 h ours ) and then immediately ch illed on ice. Bacterial cells were collected by centrifugation at 6,000 rpm at 4 C for 10 minutes The cell p elle ts were washed once in 20 ml TM B uffer (50 mM Tris HCl [ pH 7.5 ], 10 mM MgSO4 ) and stored overnight (9 h ours ) at 4 C Inner membrane vesicles were made by the French press method using the proced ure from Bhatt et. al ., 2005 (148) with a few changes The cell pellets were suspended in 3 ml TM B uffer and 20 Units of Ambion DNaseI were added Cells were passed through a Fresh press twice at 12 14,000 PSI Lysate s were then cleared of cell debris by centrifugation at 9,000 rpm (16,000 x g) in a JA 10 Beckman rotor for 10 minutes at 4 C Supernatant s were then subjected to another round of centrifugation at 9,000 rpm for 10 minutes. The resulting s upernatant s were again centrifuged at 100,000 x g at 4 C for 1.5 hours in an ultracentrifuge using a Beckman 70.1 Ti rotor. The m embrane pellet s were suspended in 1 ml cold TM B uffer using a tissue homogenizer and centrifuged again at 100,000 x g for 1.0 hour. After a final ultracentrifugation step membranes were suspended in 1 ml TM buffer using a tissue homogenizer Membrane protein concentration s were determined using a standard bicinchoninic acid (BCA) assay w ith bovine serum albumin (BSA) as the standard. Membranes were stored at 4 C for up to 36 h ours until biochemical assays were performed. S mutans The 10 ml culture s were starte d from freezer stocks and after 8 hours of growth were used to inoculate 100 ml culture s of BHI (Brain Heart Infusion) broth supplemented wit h 20 mM DL threonine, and incubated overnight at 37 C. The overnight culture s were used to inoculate 1 l iter of BHI with 20 mM DL t hreonine and

PAGE 88

88 grow n to early log phase ( OD 600 0.45 0.55 ) C ells were pelleted by centrifugation at 9,000 rpm (16,000 x g) in a JA 10 Beckman rotor for 10 minutes at 4 C Cells were w ash ed once in 10 ml B uffer A (10 mM Tris HCl [pH 6.8], 10 mM Mg acetate, in 25% sucrose ), and transferred to a 40 ml tube for a JA 20 B eckman r otor and centrifuged at 10,000 rpm (4,354 x g) for 10 minutes at 4 C The pellet was resuspend in 5 ml Buffer A and 200 l of 10 mg/ml lysozyme ( final concentration 0.3 0.4 mg/ml) and 200 l mutanolysin (10,000 U/ml, 2,000 U or 0.5 m g to tal) were added. Cells were incubated at 37 C with gentle agitation for 45 mi n utes to 1 hour to protoplast cells ( protoplast formation was monitored by Gram staining). The protoplasts were pelleted by low speed centrifugation at 6,000 rpm (4,353 x g) in a JA 20 Beckman rotor for 10 min utes at 4 C and washed twice with 10 ml Buffer A. The pellet was resuspended in 3 ml of B uffer B (50 mM Tris HCl [pH7.5] and 10 mM MgSO 4 ) and 200 ul of EDTA free protease inhibitor cockt ail (Complete Roche Tablet 25X stoc k solution), 1 0 ug/ml of DNase (50ul of DNase 1 1mg/ml stock ) and 10 g/ml RNase (50ul of RNase A 1mg/ml stock) were added Cells were lysed by passage through a French p ress at 12 14,000 PSI 2 3 times unti l cells appear ed watery. U nlysed protoplast s and cell debris were removed by centrifugation in JA 20 Beckman rotor at 10,000 rpm (12,096 x g) for 10 minutes at 4 C S upernatant s were transferred to a fresh 40 ml tube and centrifuged again. The resulting supernatant was ultra centrifuge d at 100,000 X g (45K rpm) in a SW50.1 rotor for 1.5 h ours (or overnight) at 4 C. The membrane pellet was suspend ed in 0.5 to 1.0 ml of ice cold B uffer B using a Teflon homogenizer. The protein concentration s were determined by BCA assay, with BSA used as a standard.

PAGE 89

89 ATP Hydrolysis Assay s E. coli inverted membrane vesicles ATP hydrolysis was determined by measuring the production of inorganic phosphate using a modification of the method of Fiske and Subbarow (149) All reactions were kept at 37 C, and per formed in 3 ml o f 9.1 B uffer (50 mM t ris HCl 1 mM MgCl 2 pH 9.1 ) with 120 E ach strain preparation was assayed in triplicate Reactions were started with the addition of 80 l 0.15 M A TP (0.1 M ATP, 25 mM Tri s HCl [pH 7.5] ) with a final concentration of 4 mM Reactions were stopped after 2 minutes, 5 min utes and 7 minutes by adding a 435 l aliquot of the reaction mix to 2 ml i ced Stop B uffer ( 1.3 parts H 2 O, 0.6 parts HCl /molybdate [2.5% NH 4 Mo 4 O 2 4H 2 O, 4.0 N HCl], 0.4 parts 10% SDS) and vortexing. Inorganic phosphate was measured by adding 100 l of a 1:10 dilution of Eikonogen solution (1 M NaHSO 3 0.1 M Na 2 SO 3 0.01 M 4 amino 3 hydroxyl 1 napthalenesulfonic acid) followed by incubation at room temperature f or 30 minutes The optical density at 700 nm was read with a spectropho tometer. A standard curve was generated using the follow ing concentrations of phosphate in 1 ml of reaction buffer (9.1 B uffer ) : lutions of 2 mM and 20 mM KH 2 PO 4 buffer s ) A 435 l aliquot of each standard concentration was added to 2 ml of iced Stop B uffer and processed as with samples containing membrane proteins AT P hydrolysis specific activity was calculated for each membrane sample based on the standard curve and expressed as n Mole Pi/min/mg membrane protein Data was statistically analyzed by One Test using the GraphPad Prism 4.0 program.

PAGE 90

90 S. mutans inverted membrane vesicles. ATP hydrolysis assay s with S. mutans inverted membrane vesicles were performed essentially the same as with E. coli except each reaction was performed in ATPase Buffer 6.0 (50 mM Bis Tris HCL [pH 6.0], 10 mM MgCl 2 ) with 115 g of membrane proteins F or assays with 50 M orthovanadate (P type ATPase inhibitor) membranes were incubated for 5 minutes at 37 C with the inhibitor before ATP was added to start the reaction Statistical analysis was calculated using One Multiple Comparison Post Tests, performed using the GraphPad Prism 4.0 program., Proton Motive Force Assay s The p roton motive force (PMF) of inner membrane vesicles of the E. coli YidC depl etion strain JS7131 expressing S. m utans YidC1 or YidC2 was determined by the fluorescence quenching of 9 amino 6 chloro 2 methoxyacridine (ACMA) as previously described (150) Positive controls included JS7131 rescued with E. coli YidC and the MC1060 wild type parent strain of JS7131. Negative controls included JS7131 harboring the pACYC184 vector only and the F 1 F o ATP synthase negative muta nt strain 1100 BC (151) The assay was performed by adding 500 proteins to 3 ml of MOPS B uffer (50 mM MOPS, 10 mM MgCl 2 pH 7.3) in a quar tz cuvette with a flea to stir the reaction The cuvette was placed in the fluorescence spectrometer (Photon Techn ologies International [PTI] Qu antaMaster 4), s et to an excitation of 410 nm and an emission at 49 0 nm which was recorded by the Felix32 software (PTI) provided with the spectrometer A zero baseline was recorded for 30 seconds before the addition of 15 l of 0.2 mM ACMA (final concentration of 1 M) The reaction was started 45 seconds after ACMA was added by the addition of ATP to a

PAGE 91

91 final concentration of 0.75 mM (15 l of 0. 15 M ATP 25 mM T ris HCL, [ pH 7.5 ] ). The acidification of the membrane vesicl es was determined by monitoring the fluorescence quenching of ACMA for 300 seconds, which was recorded by the Felix32 software Each trace was then combined in Adobe Photoshop As a control t he i ntegrity of membrane vesicles wa s determined by the addition of 5 l 0.1 mM nicotinamide adenine dinucleotide ( NADH ) instead of ATP to each sample and monit oring the fluorescence of ACMA. Western Blots Western blots were performed using of protein loaded on a 10% SDS PAGE (Lae mmli) gel. Proteins of interest were detected by immunoblotting using the ECL Western blot detecti on kit (Amersham Biosciences) YidC proteins were identified with antisera from rabbits immu nized with C terminal synthetic peptides f rom E. coli YidC, or S mutans YidC1 or YidC2. Western blots were also reacted with polyclonal antisera against E. coli full length YidC or polyclonal anti serum to Lep (leader peptidase) Antibodies used to recognize YidC1 and YidC2 were described in the Material and Methods of Chapter 2 The C terminal peptide (CLEKRGLHSREKKK ) antiserum against E. coli YidC was obtained from Rosemary S tuart at Marquette University. Polyclonal antiserum against Lep was provided by Ross Dalbey at Ohio State University. ATP H ydrolysis Activity of S. mutans Permeablized Whole Cells A 25 ml culture of each strain (SP22, AH374, AH378, SP13, SP14, and SP17) was grown to mid log phase (OD 600 =~ 0.5) in THYE pH 7.0 at 37 C Cells were pel leted and re suspended in 2.5 ml B uffer B (50mM Tris HCL, 10 mM MgSO 4 l of toluene was added. Cells were vortexed for 20 seconds and incubated in a 37 C water

PAGE 92

92 bath for 5 minutes They were then snap frozen in an ethano l dry ice bath and thawed at 37 C. This was repeated two times Cells were t h en pelleted and suspended in 1 ml of B uffer B and immediately used for ATP hydrolysis assays For a ssay of permeablized cells were added to 3.0 ml assay B uffer 6.0 (50 mM Bis Tris HCL [pH 6.0], 10 mM MgCl 2 ). of 0.15 M ATP was added (3 mM final concentration) Time point zero was taken approximately 15 seconds later when a 435 2 ml iced Stop B uffer ( 1.3 parts H 2 O, 0.6 parts HCl/molybdate [2.5% NH 4 Mo 4 O 2 4H 2 O, 4.0 N HCl], 0.4 parts 10% SDS) and vortexe d Another aliquot was taken at 5 minutes and again at 10 minutes After all time points were taken, tubes were removed from ice and warmed for 10 minutes at room temperature. Inorganic phosphate was measured by adding 100 l of a 1:10 dilution of Eikon ogen solution (1 M NaHSO 3 0.1 M Na 2 SO 3 0.01 M 4 amino 3 hydroxyl 1 napthalenesulfonic acid) to each tube and incubated at room temperature for 30 minutes The optical density at 700 nm was read with a spectrophotometer A standard curve was created fro m dilutions of 2 mM and 20 mM KH 2 PO 4 B uffer protein concentration was measure by BCA assay using a 1/50 dilution of permeablized cells. BSA was used as a standard. Statistical differences were determined using test. Results and Discussion Confirmation of Expression and Membrane Localization of YidC1 and YidC2 Constructs in JS7131 Western blot analysis was performed on 40 g of membrane proteins from E. c oli JS7131 grown in LB 0.2% glucose under YidC depletion conditions (Material and Methods) h arboring pACYC184 alone or encoding YidC, 247YidC1, YidC1, 247YidC2,

PAGE 93

93 YidC2 and control strains MC1060 (parent strain of JS7131) and 1100 BC (F 1 F o ATP synthas e neg ative strain) (Figure 3 3). Western blots were reacted with antisera against C terminal peptides from YidC, YidC1, and YidC2 and with a polyclonal antiserum against full length E. coli YidC that recognized the periplasmic loop included in 247YidC1 and 247 YidC2 An antibody against Lep (leader peptidase) was used as a control. As can be seen in Figure 3 3 A, the wild type copy of YidC was depleted under the assay conditions tested in strains that were producing S. mutans YidC1 and YidC2 constructs. Figure 3 3 also shows that 247YidC1 and YidC1 migrate at the predicted size of ~ 50 kDa and ~ 24 kDa respectively ( Figure 3 3 B), and 247YidC2 and YidC2 at ~ 60 kDa and ~28 kDa respectively (Figure 3 3 C) Of note both YidC1 and YidC2 appear to be process ed in the E. coli background as they are in S. mutans In Figure 3 3 D, both 247YidC1 and 247YidC2 were recognized by the YidC polyclonal with the expected sized bands at ~ 50 kDa and ~ 60 kDa respectively. Collectively these results indicate that E. col i chromosomal yidC was repressed under assay conditions and the S. mutans YidC1 and YidC2 protein constructs were produced appropriately. Restoration of Growth o f the E. coli YidC Depletion Strain JS7131 by S. mutans YidC1 and YidC2 in Broth Gro wth curves of strain JS7131 containing pACYC184 encoding either wild type YidC from E. coli or YidC1/YidC2, YidC 1, YidC2, 247YidC1, or 247YidC2 are shown in Figure 3 4 and m ean doubling times are shown in Table 3 1. As a negative control the pACYC184 vector alone was also included. Cultures were grown in LB medium containing 0.2 % glucose (Yi dC repressing conditions) or 0.2 % arabinose (condition where yidC is expressed from the chromosome by way of the araBAD promoter ). When strains w ere grown in the presen ce of 0.2% arabinose all strains gre w similarly

PAGE 94

94 regardless of which gene was expressed from pACYC184 However, when strains were grown in the presence of 0.2 % glucose, in which expression of the chromosomal copy of yidC is repress ed, a differential in gro wth was seen, with the vector only control displaying the slowest growth ( doubling time, 175.1 1.83 minutes) and the strain producing wild type YidC growing the best ( doubling time, 89.7 0.4 minutes). Expression of S. mutans y idC1 (128.3 1.0 minutes) and y idC2 (116.9 4.4 minutes) improved growth of JS7131 above the vector only control, even in the absence of the 247 periplasmic loop from E. coli YidC. Growth of JS7131 wa s improved more by 247YidC1, tha n by YidC1 without the periplasmic loop (107.1 2 .1 minutes compared to 128.3 1.0), while there was little difference between 247YidC2 and YidC2 without the periplasmic loop (112.0 1.4 minutes compared to 116.9 4.4 ). Co expression of both y idC1 and y idC2 tog ether did not show improvement over expression of y idC2 alone (113.7 1.2 minutes compared to 116.7 4.4 ) Complementation of JS7131 Growth by S. mutans YidC1 and YidC2 on Solid Media Growth on solid media, which presents a different set of challenges from growth in broth culture, of JS7131 containing plasmids encoding the S. mutans YidC1 and YidC2 constructs was also evaluated. Strains were grow n overnight in 10 ml LB medium with 0.2 % arabinose. Cells were pelleted and washed twice with 10 ml LB without sugar, before being streaked on LB a gar plates supplemented with 0.2 % ara binose or 0.2 % glucose. Figure 3 5 sho ws growth of JS7131 producing S. mutans YidC1 YidC2 or chimeric constructs on LB ag ar plates. Under these conditions only constructs producing 247YidC2 and 247Y idC1, and to a le sser extent YidC1 and YidC2 together were able to restore growth to JS7131 on LB 0.2 % glucose agar plates Given the

PAGE 95

95 results in broth cultures, this was somewhat surprising but suggests a functional requirement of the periplasmic loop of YidC that can b e overcome by co expression of YidC1 and Y idC2 Rescue of F 1 F o ATPase Activity and PMF in JS7131 by S. mutans YidC1 and YidC2 The m embrane ATPase activity in E. coli is associated primarily with the F 1 F o ATP synthase I n the absence of the F 1 F o ATPase, background ATPase activity is quite low, with roughly 300 nMol Pi/min/mg compared to 2,440 nMol Pi/min/mg in its presence (152) ATP hydrolysis is mediated by the F 1 portion of the enzyme, and is greatest under conditions when the F 1 is not attached to the F 0 (personal communication, Dr. Brian Cain). Inverted membrane vesicles were prepared from the E. coli Y i dC depletion strain JS7131 containing the various S. mutans YidC1 and YidC2 constructs expressed from a plasmid and grown under repressing conditions for yidC (LB 0.2% glucose) The ATP hydr olysis activity of the membrane vesicles w as measured in pH 9.1 Buffer, so that disassociation of the F 1 would occur and higher ATPase activity could be measured The results of the ATP hydrolysis activity assays are shown i n Figure 3 6 The YidC depletion strain JS7131 harboring the vector only had 772 84 nM ol Pi/min/mg protein of ATP hydrolysis activity. When JS7131 was complemented with E. coli YidC the specific activity was restored to 1,2 25 91 nMol Pi/Min/Mg Both YidC1 (964 64 nMol Pi/Min/Mg) and 247YidC1 ( 1,159 120 nMol Pi/min/mg) improved the ATP hydrolysis activ ity compared to the vector only. Similarly, expression of YidC2 also restored specific activity ( 1081 101 nMol Pi/Min/Mg ) which was comparable to 247YidC2 ( 1060 68 nMol Pi/Min/Mg). Therefore, in the case of ATP hydrolysis both YidC1 and YidC2 were able to complement the E. coli YidC depletion strain JS7131.

PAGE 96

96 The function of the F 1 F o ATPase is directly related to the generation of a PMF in E. coli which is generated by the translocation of protons throug h the F o channel coupled to ATP hydrolysis by the F 1 Therefore another way to evaluate the insertion and function of the F 1 F o ATP synthase is by measuring the PMF of inverted membrane vesicles There is no PMF generated in an F 1 F o ATPase mutant. The same inverted membrane vesicles used for ATP hydrolysis assays were used for PMF assays. T he results for the PMF assay s are shown in Figure 3 7 Both YidC1 and YidC2 function ed to insert and assemble components of the F 1 F o ATP synthase in E. coli and similar to the E. coli YidC, were able to restore the P MF a bove the vector only control The restoration of PMF was not improved by appending the first 247 amino acids of E. coli YidC onto either YidC1 or YidC2. To look at insertion of the individual subunits of the F o complex by the YidC proteins from S. mutans p rotease accessibility assays were performed at Ohio State University, by In these assay protoplast s from the various JS7131 strains over expressing the protein subunit of interest in the presence of a radio label, were exposed to proteinase K (PK) digestion, followed by immunoprecipitation of the protein subunits If the protein was properly inserted into the membrane it was processed by PK, producing a s maller fragment which was then visualized by phosphorimaging. The results from these experiments, which can be viewed in Dong and Palmer, et. al. 2008 Figure 4 (142) show that 247YidC1, 247 YidC2 or YidC1/YidC2 together can each function to insert nd the subunits of the F 1 F o ATP synthase T he ability of S. mutans 247 YidC1 and 247 YidC2 to insert substrates not found in S. mutans such as CyoA of cytochrome b o oxidase (94,

PAGE 97

97 153) and a derivative of phage coat protein M13 (PClep), was also observed using protease accessibility assays In these experiment s 247YidC1 247YidC2 or YidC1/ YidC2 together w ere able to insert CyoA N P2 and PClep into the membrane of E. coli Figure 6 (142) E. coli YidC can Restore Acid and Salt Tolerance to a yidC2 M utant in S mutans The yidC2 mutant of S. mutans displays a stre ss sensitive phenotype which is manifested by impaired growth at pH 5.0 or with 4% NaCl (31) The ability of E. coli YidC to com plement this phenotype was evaluated in broth culture For this purpose a chimeric gene was engineered ( see Figure 3 2 for diagram of construction) that encodes the first N terminal 50 amino acids of YidC2 including the targeting domain and SPaseII cleav age site, was appended to amino acids 59 to 548 of E. coli YidC (referred to as 50YidC) was engineered and expressed from plasmid pDL289 ( E. coli to Strept ococcal shuttle vector) under the control of the yidC2 promoter (plasmid pSP4, Table 2 1). The recombinant plasmid was then used to transform the yidC2 m utant strain AH398, resulting in strain SP01. SP01 was evaluated by growth curve under non stress (THYE pH 7.0), acid s tress (THYE pH 5.0) and osmotic stress (THYE, 4% NaCl) conditions. As s hown i n Figure 3 8, 50YidC was able to restore acid and salt tolerance to a yidC2 mutant of S. mutans Figur e 3 9 shows Western blot results with YidC1, YidC2 and YidC C terminal antibodies, indicating appropriate expression o f YidC1, YidC2 and the 50Y idC protein in the yidC2 mutant background. When yidC1 was deleted from this strain to produce SP25, mutants were barely viable and took three days to grow (data not shown). This indicates that while introduction of E.

PAGE 98

98 coli YidC into the yidC2 strain was able to complement the stress sensitivity of the yidC2 mut ant, it could not do so in the absence of yidC1 Involvement of S mutans YidC1 and YidC2 in Membrane ATPase A ctivity In order to measure membrane associated ATPase activity in S. muta ns i nverted m embrane vesicles were made by first removing the cell wall through enzyme digestion (see Materials and Methods for details) Cells were broken by passage through a French press resulting in inverted membrane vesicles which were isolated by high speed ultracentrifugation ATP hydrolysis activity was determined by measuring the production of inorganic phosphate in the presence of ATP as described above for E. coli membranes. However reactions were performed in pH 6.0 B uffer instead of 9.1, because S. mutans F ATPase functions optimally at pH 6.0. Also, it is not known if S. mutans F 1 F o ATPase displays the same properties as the E. coli F 1 F o ATPase in the presence of basic buffer. ATP hydrolysis assay s were per formed with or without 50 M orthovanadate ( a P type ATPase inhibitor) using membranes from wild type NG8, and yidC1 and mutants (Figure 3 10) There was a significant decrease in membrane associated ATPase activity in both yidC mutants (P< 0.01, by ANOVA with ultiple Comparison Test) with specific activities of 100 nMol Pi/min/mg protein in yidC1 (37% decrese) and 90 nMol Pi/min/mg protein in yidC2 (4 4% decrease), compared to NG8 membranes ( 160 nMol Pi/min/mg protein ) There was a slight but not significant decrease (13%) in NG8 specific activity i n the presence of the P type inhibitor orthovanadate with a specific activity of 140 nMol Pi/min/mg prote in. However, there was no inhibition in either of the yidC specific activities in the presence of P type inhibitor This suggests tha t the decrease seen in membrane

PAGE 99

99 associated ATPase activity in the m utants may be due in part to improper insertion of P type ATPases as well as the F type ATPase. Membrane associated ATPase assays were also per formed on a panel of strains that were designed to evaluate the function of the C terminal tails of YidC1 and YidC2 and described in Chapter 2 These included SP17, a yidC2 mutant complemented with YidC2, SP13 ( yidC2 containing YidC1C2 ) and SP14 ( yidC2 containing YidC2C1 ) Strain AH412, which has a chromosomal deletion of the C terminal tail of yidC2 and SP22 which has an Erm R marker inserted into the gtfA gene and thus serves as a control (Table 2 1 for strain details) were also included Re sults a re shown in Figure 3 11. Strain SP22 had the greatest level of specific activity with167 12 nMol Pi/min/mg protein, while strain AH412 had the lowest level of A TPase activity with 87 0.6 nMole Pi/min/mg protein. Strain AH378 showed a significant decrease (P < 0.01) in ac tivity compared to SP22 with a specific activity of 123 5 nMol Pi/min/mg protein. T his activity was significantly restored in the presence of YidC1C2 ( SP13 ) with 155 6 nMol Pi/min/mg protein (P < 0.05) Strain SP14, which contai ns YidC2C1, had similar ATPase activity to the yidC2 strain AH378 with 128 14 nMol Pi/min/mg. The positive control stra in SP17, which contained a full length wild type version of YidC2 in the gtfA locus of the yidC2 background, had a specific activity o f 136 8 nMol Pi/min/mg which was not significantly different from specific activity of 128 14 nMol Pi/min/mg protein nor was it significantly different from the positive control strain SP22 (analyzed by One iple Comparison Test). However, t his was not the complete level of complementation that would be expected for the wild type YidC2 protein. To rule out contamination of cultures and confirm the proper level of

PAGE 100

100 expression of chimeric proteins Western blot s were per formed on the mem brane samples used in these experiments. For Western blots 5 g of membranes from each sample were separated on a 12% SDS polyacrylamide gel transferred to PDVF membranes and reacted with YidC2 C term inal YidC2 TM 2/3 and YidC1 C term inal antibodies (Figure 3 12). Results show the expected reactivity with the Yid C2 C term inal and YidC1 C term inal antibodies as seen in Figure 2 6. H owever reactivity was weak in strain SP17 with the YidC2 TM 2/3 antibody (middle pa nel Figure 3 12), which recognizes a cytoplasmic loop of the YidC2 protein (Figure 2 3) A problem with expression can be ruled out s ince reactivity of the SP17 strain with the yidC2 C term inal antibody was similar to that of the SP13 strain DNA seque ncing revealed no mutation is the y idC2 gene from fresh ly grown SP17 cells The ATP hydrolysis assay was repeated using permeablized whole cells instea d of inverted membrane vesicles which are technically more difficult to prepare Res ults of ATPase assays of permeablized whole cells are shown in Figure 3 13 This time strains SP22, AH374, AH398, SP17, SP13 and SP14 were compared While the ATPase activities of permeablized cells were lower in all strains compared to the activities of inverted membrane vesicles (Figure 3 10 and 3 11) the expect ed differential in activity between the control strains was similar to previous experiments. S train SP22 (wildtype) had a specific activity of 70 8 nMol Pi/min/mg while both yidC mutants had s ignificantly less activity, AH374 ( yidC1 ) with a specific activity of 51 10 nMol Pi/min/mg ( P < 0.05, by test) and AH378 ( yidC2 ) with 30 4 nMol Pi/min/mg ( P < 0.01) of specific activity Strain SP17 ( yidC2 gtfA:: yidC2 ) had 49 7 nMol Pi/min/mg of specific activity which was significantly higher ( P < 0.05) tha n AH378. Strain SP13 ( yidC2

PAGE 101

101 gtfA:: yidC 1C2 ) also had a specific activity which was significantly higher tha n AH378 ( 42 5 nMol Pi/min/mg, P < 0.05) while SP14 ( yidC2 gtfA: : yidC2 C1 ) had a specific activity of 32 5 nMol Pi/min/mg which was comparable to the yidC2 mutant, AH378 In summary, YidC1 and YidC2 from S. mutans were expressed in the E. coli yidC depletion strain JS7131 and evaluated for their ability to complement for growth (in broth and on solid media), restore ATP hydrolysis activity and PMF associated with a functional F 1 F o ATPase enzyme, as well as to insert the a and c subunits of the F o integral membrane component of the F 1 F o ATP synthase. Both YidC1 and YidC2 were able to restore growth to JS7131 under Y idC depletion conditions in broth (Figure 3 4) H owever only the construct that contained both YidC1 and YidC2 or the chime ric vers ions of YidC1 or YidC2 (247YidC1 and 247YidC) in which the first 247 amino acid of E. c oli YidC were appended, were able to restore growth on LB 0.2% glucose agar plates ( Figure 3 5). In addition, both YidC1 and YidC2 were able to restore ATP hydrolysis activity to the YidC depletion strain JS7131 (Figure 3 6) PMF was restored by YidC1 and YidC2 as well as the chimeric versions (247YidC1 and 247YidC2), to the same extent as with wild type E. coli YidC. W hen a chimeric version of E. coli yidC encoding amino acids 59 548 of YidC was appended to S. mutans DNA incorpo rating the yidC2 promoter and encoding the N terminal first 50 amino acids of YidC2 including the signal sequence and l ip oprotein processing signal and the chimeric gene was expressed from a plasmid in the yidC2 mutant tolerance to acid and osmotic stress was restored ( Figure 3 8). T his complementation is probably the result of cooperation between YidC1 a nd the 50YidC chimeric protein, s ince 50YidC could not complement a double deletion of yidC1 and

PAGE 102

102 yidC2 Also of note, when surface adhesin P1 was evaluated in this strain, a dominant negative effect was seen, with barely a ny P1 detected on the surface of cells as ana lyzed by whole cell dot blot with a battery of monoclonal antibodies and polyclonal antiserum that recognize P1 (un published data). This suggests that the translocation machi nery involved with P1 secretion was impeded by the presence of 50YidC. As mentio ned earlier, deletion of yidC1 results in increase d anti P1 immunoreactivity and hyper adherence of S. mutans to salivary agglutinin, w hile deletion of yidC2 decreases adherence and affects only some of the monoclonal antibodies, including those that recog nize epitopes associated function This suggests that YidC2 direct ly or indirectly contributes to the translocation and maturation of P1, and that balanced expression of YidC1 and YidC2 participate in cell surface biogenesis in S. mutans. The pronounced inhibitory effect of 50YidC on P1 immunoreactivity and function suggests that it may act as a sink for critical components of the translocation machinery while not supporting transport of cell surface localized proteins in this organism The contribution of YidC1 and YidC2 to membrane associated ATP hydrolysis activity was evaluated in S. mutans using a panel of mutant s and complemented strains (Table 2 1) I nverted membrane vesicles prepared from wild type NG8, yidC1 and yidC2 were ev aluated for ATP hydrolysis activity with or without the P type inhibitor orthovanadate (Figure 3 10). Both yidC mutants displayed significantly less membrane associa ted ATPase specific activity than w ild type NG8 with only 63% of wildtype activity in yi dC1 and 56% in yi dC2 Pre incubation with orthov anadate had a modest but effect on w i ldtype ATPase activity, with 13% reduction in activity T here

PAGE 103

103 was no effect on ATP hydrolysis activity of either yidC mutant. This suggests that YidC1 and YidC2 may be involved with insertion of P type ATPases as well as F type ATPases. The a bility of chimeric YidC1 and YidC2 proteins in which the C terminal domains were swapped to restore m embrane associated ATPase activity to the yidC2 m utant was evaluated P lacing the C terminal tail of YidC2 onto YidC1 (SP13) conferred on YidC1 an ability to significantly increase ATPase activity in the yidC2 mutant. However when the C terminal tail of YidC1 was placed onto YidC2 (SP14), no compleme ntation was observed for ATPase activity (Figure 3 11 and 3 13) and growth was negatively affected further (Table 2 3) ATPase activity was also measured in the mutant and in the yidC2 mutant complemen ted with wild type YidC2 ( Figure 3 11). In this experiment ATPase activity of yidC2 C (AH412) was less than yidC2 (AH378) again indicating that the C terminal tail plays an impo rtant role in this particular function of YidC2 The complemented mutant strain SP17 was evaluated in two ways for A TP hydrolysis activity, once using inverted membranes and once using permeablized whole cells. The assay per formed w ith inverted membranes did not demonstrate full restoration of membrane ATPase activity to the yidC2 mutant, as would be expected. Upon fu rther evaluation of the membranes used in this experiment, it was noted that the SP17 sample showed decreased reactivity with a non C terminal antibody against YidC2 (Figure 3 12) This antibody recognizes an epitope within the cytoplasmic loop between tr ansmembran e domains 2 and 3 (Figure 2 3). R eactivity with the YidC2 C terminal antibody w as as expected (Figure 3 12). DNA sequencing of the yidC2 gene in SP17, prepared from the original freezer stock indicated no mutations When ATPase

PAGE 104

104 assays wer e rep eated using freshly grown permeablized whole cells (Figure 3 13) SP17 displayed improved ATP hydrolysis activity, but again not to wildtype levels. This may be due to the use of the gtfA promoter instead of the yidC2 promoter to express the yidC2 gene. Possibly the level or temporal aspects of yidC2 expression contribute to proper integration and function of YidC2 within the membrane. It was recently shown that yidC2 expressi on is upregulated under conditions that cause an envelope stress response through the LiaF S R two component system (45) It is p ossible that yidC2 expression is regulated in oth er situations and by other TCSs. Alternatively a specific l evel of YidC2 may be required to restore ATP hydrolysis activity to wildtype levels In future work,t he yidC2 gene including its own promoter shou ld be placed into the gtfA locus in the yidC2 strain, as was done for PcelB yidC2 and evalua ted for ATP hydrolysis activity to determine if the promoter makes a difference in comp lementation. It would also be informative to test the effect of expressing the yidC1 gene from the yidC2 promoter.

PAGE 105

105 Figure 3 1 Clustal W alignment of the five C terminal transmembrane domains from E. coli YidC and S. mutans YidC1 and YidC2. Predicted transmembrane domains are boxed. YidC1 and YidC2 are predicted to be lipoproteins and the SPaseII cleavage sites are underlined (113)

PAGE 106

106 Figu re 3 2 Predicted membrane topologies of E. coli YidC and S mutans YidC1 and YidC2 Predictions are based on the TMPred prediction program (A) YidC from E. coli spans the membrane six times, with the first transmembrane segment serving as an uncleavab le signal sequence, followed by a large periplasmic loop S. mutans YidC1 and YidC2 are each predicted to span the membrane five times, with an additional hydrophobic region functioning as a cleavable transme mbrane targeting sequence The predicted cleav age sites between amino acids 19 and 20 for YidC1 and amino acids 22 and 23 of YidC2 are indicated (B) Schematic illustration of the chimeric proteins used in complementation studies 50YidC is a fusion of amino acid residues 1 to 50 of YidC2 and 59 to 548 of E. coli YidC The s ignal sequence and cleavage site of YidC2 is indicated by the black box The transmembrane regions of E. coli YidC are indicated by dark grey boxes 247YidC1 is a fusion of residues 1 to 247 of YidC and 26 to 271 of YidC1 247YidC2 is a fusion of residues 1 to 247 YidC and 25 to 310 of YidC2 Each contain s the uncleavable signal sequence and large periplasmic domain of E. coli YidC appended to the five transmembrane domains (grey boxes) of S. mutans YidC1 or YidC2.

PAGE 107

107 Figure 3 3. Confirmation of appropriate production of E. coli Y idC and S. mutans Y idC1 and Y idC2 in the JS7131 YidC depletion strain by Western blot analysis. Membranes were prepared from JS7131 harboring the pACYC184 vector only or the same vector contain ing genes encoding E. coli YidC, or S. mutans 247YidC1, 247YidC2, YidC1 or YidC2. The mutant and complemented strains were grown in 0.2% glucose to repress E. coli chromosomal yidC expression. The wild type parental strain of JS7131, M C1060, and the E. coli mutant strain 1100 BC (34) that lacks a functional F 1 F 0 ATPase but has not been manipulated with regard to yidC were also evaluated. Proteins were resolved on a 10% SDS PAGE gel and the YidC homologs were revealed by immunoblot ana lysis with rabbit antisera raised against C terminal peptides of (A) E. coli YidC, (B ) S. mutans YidC1 or (C) S. mutans YidC2, as well as with polyclonal antisera against (D) full length E. coli YidC, or (E) leader peptidase (Lep), a YidC independent membr ane protein. The apparent molecular mass in kilodaltons of each protein of interest is indicated in parentheses. Asterisk indicates 247YidC2 is produced as a M r ~ 60 kDa protein.

PAGE 108

108 Figure 3 4 Growth curves of E. coli YidC depletion strain JS7131 harbouring plasmid pACYC184 encoding various YidC co nstructs. C onstructs contained within pACYC184 (vector) encoded YidC, 247YidC1, 247Yi dC2, YidC1 YidC2, YidC1 or YidC2. Growth curves were performed under yidC inducing (0.2% arabinose) and repressing co nditions (0.2% glucose) Growth curves were completed in triplicate using a Bioscreen C machine in a 100 well Bioscreen C. OD600 was recorded every 15 minutes with shaking for 10 minutes between readings.

PAGE 109

1 09 Table 3 1. Complementation of g rowth of the E. coli YidC depletion strain JS7131 with constructs encoding S. mutans YidC1 and YidC2 Doubling time (min) a Protein produced from pACYC184 0.2% glucose 0.2% arabinose None (vector only) 175.1 1.8 106.7 2.5 YidC 89. 7 0. 4 97. 8 2. 2 YidC1 128. 3 1.0 105.4 2.9 YidC2 116. 9 4.4 107.3 2. 3 247YidC1 107.1 2.1 104. 3 1.9 247YidC2 112 0 1.4 101.5 2. 3 YidC1 and YidC2 113. 7 1.2 106.1 5.8 a Mean of triplicate samples standard error. Growth in broth was determined in a 100 well plate by Bioscreen. C Machine (Labsystems, Helsinki, Finland ).

PAGE 110

110 Figure 3 5 Growth on LB agar plates of the E. coli YidC depletion strain JS7131 harboring pACYC184 encoding various YidC constructs. The t op panel shows growth of JS7131 with p ACYC184 encoding either YidC, YidC1, 247YidC1, YidC2, 247YidC2, YidC1/YidC2 or pACYC184 on 0.2% glucose (left) or 0.2% arabinose (right) The k ey is shown in bottom panel.

PAGE 111

111 Figure 3 6. ATP hydrolysis activity of E. coli Y idC depletion strain JS7131 containing pACYC184 encoding the indicated YidC constructs. ATP hydrolysis activity was determined using inner membrane vesicles prepared as in Materials and Methods by measuring production of inorganic phosphate in the presence of 120 g membrane protein s and 4 mM ATP u sing the acid molybdate method (154) Specific activity is expressed in nMol Pi/min/mg protein. Statistically significant differences are indicated, = P< 0.05 compare d to pACY184 analyzed by One Test.

PAGE 112

112 F igure 3 7 Proton motive force (PMF) of E. coli YidC depletion strains JS7131 with pACYC184 encoding the YidC constructs and the E. coli control strains indicated Inner membrane ve sicles were prepared as in Materials and Methods from E. coli 1100 BC (F 1 F o ATP mutant), MC1060 (parent stain of JS7131), JS7131 containing pACYC184 or complemented pACYC184 encoding E. coli YidC, S. mutans 247YidC1, YidC1, 247YidC2, and YidC2. Five hundred micrograms of membrane proteins were used to analyze PMF, through quenchi ng of ACMA monitored with a fluorescence spectrometer The inte grity of membrane vesicles was confirmed by adding NADH to each sample and monitoring the fluo rescence of ACMA (not shown).

PAGE 113

113 Fig ure 3 8 Growth curves of S. mutans wil d type NG8 yidC1 and with chimeric E. coli 50YidC (YidC2 1 50 YidC 59 548 ) under non stress acid and osmotic stress conditions (A) Non stress, THYE pH 7.0 (B) Acid stress, THYE pH 5.0. (C) Osmotic stress, THYE 4% NaCl Growth curves were completed in triplicate using a Bioscreen C machine (Labsystems, Helsinki, Finland ).

PAGE 114

114 Figure 3 9 Western blot of whole cell lysates of S. mutans wild type, yidC1 yidC2 and mutant with chimeric E. coli YidC (YidC2 1 50 YidC 59 548 ) Proteins were resolved on 10% SDS polyacrylamide gels and identified by reactivity with rabbit antisera raised against C terminal peptides of S. mutans YidC1 or YidC2 and E. coli YidC.

PAGE 115

115 Figure 3 10. Membrane associated ATPase activity in S. mutans NG8, yidC1 and yidC2 strains with or without the P type ATPase inhib itor o rthovanadate Dark grey bars represent ATPase activity without orthovanadate Light grey bars represent specific activity after membranes were incubated for 5 minutes with 50 M orthovanadate Specific activity is expressed as nMol Pi/min/ mg of membrane proteins Error bars represent standard deviation of the average specific activity resulting from dupli cate assays. indicates a significant difference compar ed to NG8 total ATPase activity, P value < 0.01. Statistica l analysis was performed using O ne way ANOVA with Bo Multiple Comparison Post T est.

PAGE 116

116 Figure 3 11 Membrane associated ATP hydrolysis activity of S. mutans wildtype and yidC2 mutant strains expressing various YidC constructs Mean specific activity is expressed as nMol Pi/min/mg membrane protein calculated based on duplicate samples from three different time points S tandard deviation is indicated by error bars Genotypes of strains are as follows: SP22 gtfA::erm AH378 yidC2 SP17 yidC2 gtfA::yidC2 AH412 yidC2 C, SP13 yidC2 gtfA::yidC 1C2 and SP14 yidC2 gtfA::yidC2 C1. Statistically significant differences are based on One way ANOVA followed by Tu Comparison Test, and are indicted; = P < 0.05 ; = P < 0.01 ; ** = P < 0.001 compared to SP22, or = P < 0.05 compared to AH378.

PAGE 117

117 Figure 3 12 Western blot s of membrane samples prepared from S mut ans strains used in ATP hydrolysis assays (Figure 3 11) For each indicated strain use d in the ATPase assay (Figure 3 11 ) 5 g of membrane protein s were loaded on a 12% SDS PAGE gel. Proteins were transferred to PDVF membranes and reacted with the indi cated affinity purified antibodies Bottom panel shows colloidal gold protein stain. indicates a possible unprocessed versi on of the chimeric YidC2 C1 protein. Genotypes o f strains are as follows: SP22 gtfA::erm AH378 yidC2 SP17 yidC2 gtfA::yidC2 AH412 yidC2 C, SP13 yidC2 gtfA::yidC 1C2 and SP14 yidC2 gtfA::yidC2 C1.

PAGE 118

118 Figure 3 13. ATPase activity of S. mutans permeablized whole cells Assay was performed in triplicate. Average specific activity is expressed in nMol Pi/min/mg whole cell protein, with error bars representing the standard deviation of the mean of triplicate assays Genot ypes of strains are as follows: SP22 gtfA::erm AH378 yidC2 SP17 yidC2 gtfA::yidC2 AH412 yidC2 C, SP13 yidC2 gtfA::yidC 1C2 and SP14 yidC2 gtfA::yid C2 C1. Statistically significant differences are indicted = P < 0.05 and ** = P < 0.01 compared to SP22, or P < 0.05 compared to AH378. Significant test.

PAGE 119

119 CHAPTER 4 DIFFERENCES IN MEMBRANE PROTEIN COMPLEXES BETWEEN W ILD TYPE AND YIDC1 AND YIDC2 MU T A N T S USING BN PAGE Rationale for Study Blue Native polyacrylamide gel e lectrophoresis (BN PAGE) was originally developed to separate enzymatically activ e membrane proteins, such as respirat ory chain complexes located wit hin the mitochondrial membrane (155) It has also been widely used to study chloroplast membrane protein complexes such as the light harvesting chlorophyll binding proteins and t win arginine t r anslocation pathway (156) (157) T he BN PAGE technique uses gentle non ionic detergents to solublize proteins in their native conformations under non denaturing conditions Once membr ane proteins are solublized in detergent micel les, the anionic dye Coomassie Brilliant B lue G 250 is added to the protein samples The dye binds to the proteins causing a charge shift which allows for separation of pro tein complexes based on size i n a gradient Blue N ative polyacrylamide gel. Protein complexes can be further analyzed by in ge l en zyme assays, Western blot, mass s pec trometric analysis, or by second dimension PAGE, either 2D BN/BN PAGE or 2D BN/SDS PAGE. BN PAGE combined with second dimension SDS PAGE is a powerful technique that allows protein subunits of complexes to be sepa rated in a second d imension such that individual protein s can be isolated [ reviewed in (158 159) ] This technique can also be combi ned with mass s pec trometric analysis for applications in bacterial pro teomics [ reviewed in (160) ] This technique was successfully applied to inner membrane fractions from E. coli identifying 43 distinct protein complexes of the inner membrane, including the F 1 F o ATP synthase subunits, as well as YidC, YajC, SecA, and a number

PAGE 120

120 of transport complexes (161) Another study used 2D BN/SDS PAGE in conjunction wit h MALDI TOF MS (matrix assisted laser desorption/ionization time of flight mass spectrometry) to analyze the membrane proteome of the green sulfur bacterium Chlorobium tepidum (162) This group was able to identify 63 membrane proteins out of 120 t otal proteins isolated including membrane proteins containing up to 14 transmembrane domains In a stud y by Tsirogianni the composition of a phenol inducible multi enzyme complex from a Pseudomonas species was identified using BN PAGE coupled with hig h throughput mass spectrometry (163) BN PAGE can also be adapted to determine the molecular mass es of membrane proteins as well as the oligomeric states of complexes, as was demonstrated for several well characterized bacterial membrane tr ansport proteins by Heuberger (164) Of note this group found that membrane proteins migrate differently than the soluble proteins used as standards, and that by multiplying the predicted molecular mass based on am ino acid sequence by a factor of 1.8 the apparent mass by BN PAGE of membrane proteins can be estimated It is clear from the variety of studies using 2D BN/SDS PAGE, that this technique can be a useful proteomics tool In addition it has application s in determining protein complex oligomeric states as well as the potential to discover new protein protein interactions within the membranes and the cytoplasm s of cells. I n the studies outlined in this C hapter 2D BN/SDS PAGE followed by Western blot, was u sed to gain insight into potential protein protein interactions between YidC1, YidC2 and SecY from S mutans Also, o ne dimensional BN PAGE combined with LC MS/MS (liquid chromatography mass spectrometry/mass spectrometry) was used to begin to evaluate

PAGE 121

121 di fferences in mem brane protein complexes between wild type and y idC1 and y idC2 mutants in S mutans in order to begin to understand the function of the YidC/Oxa 1/Alb3 family of proteins in s treptococci. Materials and Methods Membrane Fractions and DDM Solublization Isolating cytoplasmic membranes from gram positive bacteria can be complicated because the cell wall can make cell lysis difficult For this reason cells have first been converted to protoplasts, by digesting the cell wall with lysozyme and mutanolysin Chassy found that growth in 20 mM DL threonine made cells more susceptible to lysozyme treatment by interfering with lysine incorporation in to the cell wall and inhi biting cross link formation (165) He also found that s treptococci grown in Brain Heart Infusion broth provided better lysis of proplasts tha n those grown in Todd Hewitt Broth S. mutans c ells in this study were grown in Brain Heart Infusion broth (BHI, Difco) supplemented with 20 mM DL t hreonine at 37 C in a standard incubator without aeration M embranes from S mutans wild type strain NG8 were compared with those from AH374 (NG8, yidC1 ::E r m) and AH398 (NG8, yidC2 ::E r m) For each strain a 1 ml freezer vial (25% g lycerol frozen back in mid log) was used to inoculate 9 ml med ia in a 15 ml conical tube After approximately 8 hours of growth this culture was used to inoculate a 100 ml bottle of pre warmed media and grow n overnight In the morning the 100 ml overni ght culture was used to inoculate 900 ml of pre wa rmed media in a 1 liter flask. T he 1 liter culture was grown until it reached an OD 600 of 0.5 to 0.6 (mid log phase) and the flask was placed in an ice bath for at least 10 minutes to stop growth

PAGE 122

122 Each cultu re was then split into three 500 ml Beckman centrifuge tubes and centrifuged at 9,000 rpm (16,000 x g) in a JA 10 Beckman rotor for 10 minutes at 4 C to pellet the cells. Bacterial c ells were re suspended in 10 ml B uf fer A (10 mM Tris HCl [pH 6.8], 10 mM Mg acetate, in 25% suc rose) and transferred to a 40 ml Beckman tube followed by centrifugation at 10,000 rpm (12,074 x g) in a JA 20 Beckman rotor for 10 minutes. The c ell pellet s were stored at 20 C until needed. F reezing the cells before lysozyme and mutanolysin treatment resulted in more efficient protoplast formation. To prepare protoplasts the cell pellet s were thawed and suspended in 5 ml Buffer A to which lysozyme (10 mg/ml stock) and 2,000 U mut anolysin (200ul of a 10,000 U/ml stock, Sigma # M9901) were added. Cells were incubated at 37 C with gentle agitation for approximately 1.5 hours and monitored by Gram stain until ~95% of cells were protoplasts (gram negative) The p rotoplasts were pelleted by low speed centrifugation at 6,000 rpm (4,353 x g) in a JA 20 Beckman rotor for 10 min utes at 4 C They w ere then washed twice with 10 ml B uffer A. The p rotoplast p ellet s were sus pended in 3 ml Membrane B uffer (50 mM Tris HCL [pH7.5] and 10 mM MgSO 4 ) and of EDTA free prot ease inhib itor cocktail (Complete Tablets, 25 x stock solution, Roche ), and of both DNase 1 (Sigma, D4527) and RNase A (Sigma, R5125) were added. P rotoplasts were lysed by sonicating for 3 cycles, 15 seconds each ( with cooling on ice between cycles ) on setting 7 of a Fisher Scientific Sonic dismembrator 100 C ell lysis was confirmed by G ram stain U nlysed protoplast s and cell debris was removed by centrifu gation at 6 ,000 rpm ( 4,347 x g) in a JA 20 Beckman rotor for 10 minutes at 4 C The s uper natant s containing membranes were transferred to Beckman 5 ml ( 13 x 51 mm ) tube s for a SW 50.1 Beckman rotor. The m embranes were

PAGE 123

123 pelleted by ce ntrifugation at 100,000 X g (45,000 rpm) for 1.5 hours (or overnight) at 4C The resulting m embrane pellet s were suspended of NativePAGE Sample B uffer (50 mM BisTris HCL [7.2], 50 mM NaC l 10% glycerol, 0.001% Ponceau S ) using a teflon homogenizer. P rotein concentrat ion s were determined by BCA assay with a 1/500 dilution of membranes A 1 m g/ml s tock solution of bovine serum albumin (BSA) was used to make a standard curve Each membrane preparation was adjust e d to a protein concentration of 10 For BN PAGE m embrane protein s were solublized with the non ionic detergent dodecyl maltoside (DDM ) For the solublization step of membrane proteins were combined with 4 l of 20% DDM (1% of final volume) Samples were briefly vortexed and placed on ice for 15 minutes with intermittent vortexing every 5 minutes Samples were th en ultracentrifuged at 100,000 x g ( 4 C ) for 15 minute s using a TLA 55 Beckman rotor in a Beckman Optima TLX tabletop ultracentrifuge Supernatant s were saved as the solublized fraction s Aliquots were frozen at 20 C and used later in BN PAGE experiment s. Blue Native Polyacrylamide Gel Electrophoresis The BN PAGE protocol used in these experiments was adapted from the Invitrogen NativePAGE Novex User Manual, which is based on the original protocol developed by Sch gger and von Jagnow, (155, 166) Accordingly, solublized membrane proteins were combined with 5% Cooma s sie Blue G 250 solution (5% Coomassie Blue G 250 in 500 mM 6 aminohexanoic acid) at a deter gent/dye ratio of 8 (gram/gram), which gives proteins a net negative charge allowing for separation based on size in a polyacrylamide gel BN PAGE was performed using the Invitrogen XCell SureLock Cell system. To prepare samples for electrophoresis, 80 l of

PAGE 124

124 solublized membrane proteins ( 1% DDM ) w ere combined with 2 l of 5% Coomassie Blue G 250 solution Then 7.5 of each sample was loaded on a 10 well 3 12% Invitrogen NativePAGE Novex Bis Tris gel The Invitrogen Native Mark Unstained Protein Marker (5 l ) was used as a standard For electro p horesis, 600 ml of cold NativeP uffer (50 mM Bis Tris, 50 mM Tricine, pH 6.8) was poured i nto the lower chamber and 200 ml of cold NativePAGE Dark Cathode Buffer ( 50 mM Bis Tris, 50 mM Tricine, 0.02% Cooma s sie Blue G 250 ) was poured in the top c hamber Electrophoresis was performed in a cold room at 4 C with t he XCell surelock cell connected to a PowerPac Basic Power Supply from BioRad set at 150 volts Once the dye front had reached approximately 1/3 of the way to the bottom of the gel after ~45 minutes the dark blue cathode buffer was exchanged for NativePAGE Light Cathode Buffer (50 mM Bis Tris, 50 mM Tricine, 0.002% Coomassie Blue G 250) and continued to run at 250 volts until the dye front reached the bottom (~50 minutes) BN PAGE gels were then further processed with one of the following protocols; Coomassie Blue R 250 stain, BioRad Silver Stain Plus K it BioRad Bio Safe Coomassie Blue for Mass spec trometric analysis, t ransfer to PDVF membrane for Wes tern blot analysis, or 2nd D imension SDS PAGE ( p rotocols are described below) Protein Staining of Blue Native Polyacrylamide Gels Coomas s ie Brilliant Blue R 250 s tain All membrane samples analyzed by BN PAGE were stained with Coomassie Blue R 250 as a quality control step to ensure that the membrane protein banding patterns were cons istent between each experiment Coomassie Blue R 250 stain has a limit of detection between 50 to 100 ng of protein When more sensitive stain ing was needed gels were s ilver stained (BioRad Silver Stain Plus Kit, below) After the BN PAGE run, gel were placed in approximately 100 ml

PAGE 125

125 F ix ative S olution ( 40% methanol, 10 % acetic acid ) and microwave d on high fo r 45 seconds The Gel s were then placed on an orbital shaker for 15 30 minutes at ro om temperature Next, Fixative S olution was decanted and 100 ml of Coomassie R 250 Stain (0.02% Coomas s ie R 250 in 30% methanol and 10% acetic acid) was added and microwave d on high for 45 sec onds followed by incubation at room temperature on an orbital s haker for 15 to 30 minutes Coomassie R 250 stain was removed and the g el s were placed in 100 ml of Destain Solution (8% acetic acid solution) and microwaved again on high for 45 seconds Gel s were placed on an orbital shaker until the desired background was reached Gels were stored in water until they were dried or photographed BioR ad Silver Stain Plus If a more sensitive protein stain was desired gels were silver stained This method has a limit of detection of ~2 ng of protei n per band T he BioRad Silver Stain Plus Kit following deviation s Immediately after BN PAGE ge ls were placed in a glass bowl (washed previously with 50% nitric acid) with 200 ml of Fixative Enha ncer Solution (50 % methanol, 10% acetic acid, 10 % Fixative Enhancer, 30% distilled de ionized water) and placed on an orbital shaker overnight at room temperature. In the morning Fixative Enhancer Solution was decanted, and gels were rinse d twice in 200 m l distilled de ionized water for 10 minutes on an orbital shaker During the last wash step the Silver Stain Plus solution was made according to product directions. A Teflon stir bar was added to a glass beaker ( both cleaned with 50% nitric acid solution ) on a stir plate and 35 ml of distilled de ionized water was added and the stir plate turned on To this, 5.0 ml of each of the following were added in this order; Silver Complex Solution, Reduction

PAGE 126

126 Moderator Solution, and Image Development Reagent The m ixture was stir red for approximately 30 seconds more and then immediately added to the glass bowl with the gel to be stained and placed on an orbital shaker at room temperature to develop Gels were constantly monitored for stain development and once desired stain ing was reached the Silver Stain Plus s olution was decanted and the gel was place in 5% acetic acid for 15 minutes to stop development Gels were stored in water until photographed and dried. BioRad Bio Safe Coomassie S tain Gels that were intended for Mass S pec trometry analysis were stained using Bio Safe Coomassie from BioRad, according I mmediately following electrophoresis NativePAGE gels were washed a to tal of 3 times for 5 minutes each with 200 ml distilled de ionized water During the wash steps gels were placed on an orbital shaker at room temperature After the water from the final wash step was decanted, enough Bio Safe Coomassie was added to cover each gel The gels were incubated for at least 1 hour in the stain on an orbital shaker After staining, gels were de s tained in 200 ml of distilled de ionized water for at least 30 minutes or until desired background was reached Next, gels were place d on a glass plate on a light box and bands of interest were excised using a r azor blade and place d in a 1.5 ml E ppendorf tube on ice Samples were either immediately taken to the University of Florida ICBR Protein Core for LC MS/MS analysis or frozen at 20 C and taken to the ICBR Protein Core at a later time.

PAGE 127

127 In Gel Trypsin Digestion and LC MS/MS (University of Florida, ICBR Core Facility Protocols ) Destain and SDS removal The gel slice containing the band of interest was 50% a ceton itrile solution while vortexing 2 times for 15 min utes The w ash solution was removed and the gel slice was dehydrated with acetonitrile (ACN) ACN was removed and the gel slice was rehydrated by incub ation for 5 minutes in 50 50 mM ammonium bicar bonate buffer ( NH 4 HCO 3 ) An equal amo unt of ACN was added (1:1 ratio of ammonium bicarbonate to ACN), and the sample was vortexed for 15 minutes of 50:50 ACN to 50 mM ammonium bicarbonate was added and the sample was vortexed for 15 minutes more to completely de stain the gel slice Reduction, alkylation & trypsin digest After the de stain procedure, the wash was remove d and the gel slice was dried in a S peedvac for 10 15 minutes (samples may have been stored at 4 C at this point) The s ample was r e fresh 45 mM dithiothreitol (DTT), 50 mM ammonium bicarbonate at 55 C for 45 minutes Samples were allowed to come to room temperature and DTT solution was removed, then fres h 100 mM iodoactamide (IAA), then incubated in the dark at room temperature for 45 minutes The IAA was removed and the gel slice 50% ACN/50 mM ammonium bicarbonate by vortexing on low 3 times for 3 minutes each The w ash was removed and the gel p iece w as completely dried in a S peedvac (15 minutes). The s ample may have been stored at 4 C If not, the samp le was chilled in an ice bucket and enough Trypsin S olution (12.5 ng/ul trypsin in 50 mM ammonium bicarbonate ) was added to cover the gel piece The s ample was incubated on ice for 10 minutes Excess Try psin S olution was removed and the gel was

PAGE 128

128 overlaid with 30 of 50 mM ammonium bicarbonate Trypsin digest ion was performed at 37 C for 12 to 16 hours. Gel e xtraction After trypsin digestion, the supernatant was rem oved and place d in a clean 1.5 ml tube 80:20 ACN:water containing 0.1% fo r mic acid was added to the gel piece, and vortexed for 10 15 minutes Buffer was removed and added to the previous supernatant. The g el piece wa s then dried in a S pee dvac and resuspended in 15 Loading B uffer (3% ACN, 0.1% acetic acid 0.01% TFA ) T he sample was sonicated in a water bath for 10 minutes, gently agitated on a vortex for 1 minute then centrifuged at high speed for 15 minutes. The r esulting supernatant was added to the other supernatant, and then analyzed by LC MS/MS or stored at 4 C. LC MS/MS The enzymatically digested samples were injected onto a capillary trap (LC Packings PepMap) and desalted for 5 min utes with a flow rate 10 l /min of 0.1% v/v acetic acid. The samples were loaded onto an LC Packing C18 Pep Map HPLC column The elution gradient of the HPLC column started at 3% Solvent A, 97% Solvent B and finished at 60% Solvent A, 40% S olvent B for 6 0 m in utes for protein identification Solvent A consisted of 0.1% v/v acetic acid, 3% v/v ACN, and 96.9% v/v H 2 O Solvent B consisted of 0.1% v/v acetic acid, 96.9% v/v ACN, and 3% v/v H 2 O LC MS/MS analysis was carried out on a hybrid quadrupole TOF mass spectrometer (QSTAR, Applied Biosystems, Framingham, MA). The focusing potential and ion spray voltage was set to 275 V and 2600 V, respectively. The information dependent acquisition (IDA) mode of operation was employed in which a survey scan from m/z 4 00 1200 was acquired followed by collision induced dissociation of the three most

PAGE 129

129 intense ions Survey and MS/MS spectra for each IDA cycle were accumulated for 1 and 3 s econds respectively. Protein search a lgorithm Tandem mass spectra were extracted b y ABI Analyst version 1.1 All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK; version 2.0.01) Mascot was set up to search the NCBInr database assuming the digestion enzyme was trypsin Mascot was searched with a fragment ion mass tolerance of 0.30 Da and a parent ion tolerance of 0.30 Da Io doacetamide derivative of Cys, deamidation of Asn and Gln, oxidation of Met, are specified in Mascot as variable modifications Scaffold (version Scaffold 01 06 03, Proteome Soft ware Inc., Po rtland, OR) was used to validate MS/MS based peptide and protein identificatio ns. Peptide identifications were accepted if they could be established at greater than 95.0% probability as specified by the Peptide Prophet algorithm (167) Protein identifications were accepted if they could be established at greater than 99.0% probability and contain ed at least two identifie d unique peptides. Protein probabilities were assigned by the Protein Prophet algorithm (168) BN PAGE with Western B lotting In order to determine the locations of YidC1, YidC2 and SecY in membrane protein complexes separated by BN PAGE, proteins were transferred to PDVF membranes an d Western blots were performed with affinity purified antibodies (see Material and Methods, Chapter 2) against YidC1, YidC2, and SecY The following protocol was used to transfer membrane protein complexes to PDVF membranes a fter the BN PAGE run Blue N a tive g els were incubate d in 0.1% SDS, 20 mM T r icine, 7.5 mM I midazole ( pH 7.0 ) for 10 minutes at room temperature on an orbital shaker PDVF membrane s were hydrated in 100% methanol for ~30 seconds and then rinse d in water

PAGE 130

130 for 2 minutes on an orbital shak er PDVF membranes were then equilibrate d in Transfer B uffer (20 mM Tricine, 7.5 mM I midazole [pH 7.0]) for 10 minutes Proteins were blotted using the Hoefer Mighty Small TE 22 Mini Transfer Tank system T he Blue N ative gel was sandwiched between blotting sponges and blotting paper with the PDVF membrane placed on top of the gel Once assembled, the cassette was placed in the tran sfer tank filled with Transfer B uffer, and electrophoresis took place at 50 volts for 45 minutes at room temper ature Once transfer was complete proteins were fix ed to the PDVF membrane in 20 ml 8% acetic acid solution for 15 minutes and rinse d in water to remove acid The membrane was block ed in PBS 0.3% Tween 20 5% skim milk for at least 1 hour or overnight at 4 C on an orbital shaker. Western blots were process ed and developed following the Amersham (GE Healthcare) ECL protocol. Two Dimensional BN/ SDS PAGE Second dimension Tricine SDS/PAGE Lanes cut from first dimension BN PAGE gels were subjected to a r educing and alkylating step following the protocol from Nat ivePAGE Novex Bis Tris Gel System User Manual Briefly each ex cised lane was placed in a 15 ml conical tube with 5 ml Reducing Solution (1 x NuPAGE LDS Sample Buffer with 50 mM DTT ) The t ube was placed on an orbital shaker at room temperature for 15 to 30 m inutes. Reducing solution was decanted and replaced with 5 m l of Alkylating Solution (1 x NuPAGE LDS Sample Buffer with 50 mM DMA (N,N dimethylacrylamide) and i ncubated at room temperature on an orbital shaker for 15 to 30 minutes Alkylating Sol ution was decanted and 5 ml of Quenching Solution (1 x NuPAGE LDS Sample Buffer, 5 mM DTT, and 20% ethanol) was added and incubated at room temperature for 15 minutes The g el strip was then place d on top o f a 10% Tricine SDS PAGE gel to run in the second dimension

PAGE 131

131 T ricine SDS PAGE was per formed essentially as described by Hermann s Nature Protocol from 2006 (169) Tricine g el s were cast using the BioRad Mini Protean II casting stands, with 12.5 cm x 6.5 cm x 1.5 mm glass plates with a prep well comb Gels were made with a solution containing 1 M T ris HCl (pH 8.45), 0.1% SDS ( sodium dodecyl sulfate ) 10% glycerol, 10% acrylamide (49.5% T, 3% C mixture), 0.05% ammonium per sulfate and 0.0005% TEMED Gels were made t he day before or the same day that 2D SDS PAGE was per formed For the second dimension, a T ricine gel was placed in a B ioRad Mini Protean II Cell and Cathode B uffer (100 mM Tris, 100 mM Tricine, and 0.1% SDS [~pH 8.25]) was poured into the upper buffer chamber Equilibrated gel strips (re duced, alkylated and quenched) were then placed in t he prep wel of 1 x LDS (lithium dodecyl sulfate) sample buffer was applied around the gel strip Next, Anode B uffer (100 mM Tris HCl [pH 8.9]) was poured into the lower bu ffer chamber and the Mini Protean II Cell was connected to a BioRad PowerPac Basic Power Supply set at 150 volts. The gel was run for ~ 1 hour or until the dye front had reach ed the bottom of the gel. After electrophoresis, gels were either transferred to PDVF membranes for Western blotting (see below) or stained using the BioRad Silver Stain Plus Kit, as with 1 st dimension BN PAGE gels (described above). Western blots of second dimension Tricine SDS PAGE Second dimension gels were transferred to P DVF membranes and Western blots were p er formed as described in Chapter 2 under Material and Methods under Western blotting For these experiments membranes were probed with affinity purified YidC1 (1:2000), YidC2

PAGE 132

132 C terminal (1:6000), and SecY (1:500) antibodies. Antibodies are described in detail in Chapter 2 in Materials and Methods under Synthetic Peptide Antiserum. GAPDH Assays with Whole Cells GAPDH activ i ty was determined using cells grown overnight in THYE Cell s were pelleted and resuspended to an OD 600 of 1.0 in 25mM Tris HCL [ pH 7.5], 5mM EDTA For assay 1 ml of ce ll suspension was placed in an E ppendorf tube, spu n down and of GAPDH Assay B uffer ( 40 mM Triethanolamine 50mM Na 2 HPO 4 5mM EDTA, pH 8.6 ) The OD 600 for each strain was read and recorded again before the assay was performed. For each assay 7 of gly ceraldyhyde 3 p hosphate (50 mg/ml Sigma ) was added along with 100 of 10 mM NAD+ at which point a timer was star t ed and reactions were place d in a 37 C water bath for 30 minutes. After incubation cells were spun at 4 C for 2 min utes and OD 340 was read Assay was performed in triplicate for each strain Results are expressed as OD 340 /OD 600 Results and Discussion Changes in Protein Complexes in YidC Mutant Membranes were Visible by First Dimension BN PAGE Membrane p roteins were isolated from wild type NG8, and mutant strains of S. mutans The membrane proteins were solublized using 1% DDM (do decylmaltoside), a non ionic detergent that preserves membrane protein complexes. S amples were compared using the B N PAGE technique which separates native membrane protein complexes based on size in a polyacrylamide gel. Sh own in Figure 4 1 is the resul ting banding patterns from wild type NG8, yidC1 and yidC2 solublized membrane protein complexes after staining with Bi

PAGE 133

133 Coomas s ie Blue G 250 stain There were visible difference s in banding patterns between the yidC mutants and NG8 (Figure 4 1 and 4 2 ) p articularly in high m olecular weight complexes of NG8 between 480 and 720 kDa whic h were not visible in either yidC1 or yidC2 mutant s There wa s also an apparent lack of a band at approxi mately 300 kDa in both yidC mutants ( Figure 4 1) Additionally, there were a number of bands present in the mutants in lower mol ecular weight complexes that were not present or were less pronounce d in NG8 These migrate d roughly between ~ 146 and ~ 60 kDa in F igure 4 1, 4 2 and 4 5 S hifts in the location of bands could indicate that in the mutant membranes some protein complexes were disrupted or perturbed, there could have be en a change in expression of certain proteins or there was a defect in the ass embly of multimeric protein complexes YidC1, YidC2 and SecY Co Migrate in High Molecular Weight Complexes In an attempt to determine whether YidC1, YidC2 and SecY are present in a complex together in S. mutans BN PAGE was used in conjunction with Weste rn b lot ting Figure 4 3 shows a large streak of reactivity with the YidC1 and YidC2 antibodies suggesting that both YidC1 and YidC2 are located in a number of high molecular weight complexes The pattern s of reactivity are not identical with YidC1 ranging from the top of the gel to approximately 300 kDa, and YidC2 from the top of the gel to approximately 500 kDa There appears to be an extension in the location of YidC1 in the yidC2 mutant sample, with the re activity extending down to approximatel y 200 kDa. The SecY antibody reactivity was weak; however there was also a long streak of reactivity seen starting at the top of the gel and extending to approximately

PAGE 134

134 300 kDa with a pronounced band at ~146 kDa in the NG8 and yidC1 lanes, that was abse nt in the yidC2 mutant. Ep itopes may be masked in Western blots of first dimension Blue Native polyacrylamide gel s if the epitope is not exposed in the native protein, or by other proteins located in the same complex Combining BN PAGE with second dimension SDS PAGE followed by Western blot can separate protein s from complexes in the second dimension and allow for better epitope recognition (160) Figure 4 4 shows the res ults of second dimension SD S PAGE Western blots with YidC1, YidC2, and SecY antibodies As in the 1D BN PAGE Western blots (Figure 4 3) YidC1, YidC2 and SecY are located in a streak of unresolved high molecular weight complexes The streak observed with the YidC1 antibody migr ated at ~ 24 kDa and was only seen in the NG8 and y idC2 samples. Likewise, the streak observed with the YidC2 antibody migrated at ~ 28 kDa and was only seen in NG8 and the yidC1 samples. Second dimension Western blots with the SecY antibody produced reactivity in three different locations corresponding to three different sizes For the NG8 sample there was a ~ 75 kDa spot corresponding to the ~146 kDa band in first dimension Blue Native gel, a horizontal streak running at ~ 37 kDa, and a horizontal s treak at ~20 kDa In the yidC1 mutant sample there were two SecY species, a ~ 75 kDa spot and a 37 kDa horizontal streak as in NG8 There was only faint reactivity in the yidC2 mutant sample with the SecY antibody, corresponding to the 37 kDa streak on ly The ~20 kDa streak that was present in the NG8 sample was absent from the yidC1 and yidC2 mutant samples. SecY has a predicted Molecular Weight of ~ 48 kDa in S. mutans however runs at about ~ 37 kDa on SDS PAGE gels. Others have found that SecY f rom E. coli runs

PAGE 135

135 aberrantly on SDS PAGE gels, migrating at about ~ 35 kDa with a predicted Molecular Mass of ~ 48 kDa (170 171) It is not uncommon for very hydrophobic proteins to run aberr antly on SDS PAGE gels. T hey bind more SDS and therefore migrate faster (172) The ~ 75 kDa species in S mutans could be a dimeric f orm or possibly an unresolved heterotrimeric SecYEG complex, which would run at about that size. The oligomeric state of SecY is somewhat of a controversy in the protein translocation field, [ reviewed in (52) ] It can clearly functi on as a monomer; however some have reported it as a dimer or in higher oligomeric states when i t is over expressed (171, 173) SecY was identified in a 400 kDa complex containing SecYEG and FtsY, i n E. coli using BN PAGE combined with Western blot with a SecY antibody (174) SecY was also found as a 230 kDa co mplex containing a SecYEG dimer, as we ll as 440 kDa and 880 kDa complexes in E. coli when S ecY was crosslinked as a dimer and co expressed with SecA dimers (175) The ~ 20 kDa moiety i n the NG8 Western in Figure 4 4 probably corresponds to a breakdown product of SecY, which is kno wn for its susceptibility to the protease FtsH (176) In studies by van Bloois (177) and Price (139) YidC was able to be cross linked to FtsH, HtlC and HtlK, suggesting that in E. coli YidC plays a role in quality control and maturation of membrane proteins. The decrease of SecY observed in the y idC2 mutant may indicate that YidC2 plays a regulatory role in FtsH activity or in the activity of other cellular proteases. A study in E. c oli that involvement in membrane protein biogenesis found that the amount of SecY was reduced in the membrane when FtsY was depleted (178) I t is possible that FtsY could also be affected by deletion of yidC2 in S. mutans Additionally, SecE is also know n to

PAGE 136

136 require YidC for insertion in E. coli (179) S everal studies have found that decreasing the level of SecE resu lts in reduced SecY in the membrane (127, 180) Therefore, YidC2 could play a direct role in the insertion of SecY or in the insertion or regulation of other proteins that effect SecY stability. Furthermore, SecY is located in an operon with ribosoma l proteins, which could be down regulated in the absence of YidC2 due to a cellular stress response. In microarray studies of an ffh mutant in S. mutans (which has a similar growth and stress sensitive phenotype to the yidC2 mutant) a number of proteases were up regulated including; HtrA, putative ATP dependent Clp protease (SMU.1672), a Zn dependent protease (S MU.1438c), and class III stress response related ATP dependent Clp protease (SMU.2029) (75) In this same study there were also a number of down regulated proteins involved in translation including several ribosomal prote ins, putative translation initiation factor IF 1 (SMU.2004), 50S ribosomal protein s L7/L12, and L33. In S. mutans the s ecY gene ( SMU.2006 ) is located upstream of translation initiation factor IF 1 (SMU.2004 ) which was down regulated in the ffh mutant and could possibly be coordinately regulated. There is no microarray data yet available for the yidC1 or yidC2 mutants in S. mutans However, considering the similar phenotype of the ffh and yidC2 mutants, it is likely that many of the s ame pathway s involved in stress tolerance would be affected. Inte restingly, the smear recognized by YidC1 was longer in the y idC2 mutant for both Westerns of first dimension and second dimension gels, indicating that YidC1 may be located in different complexes i n the absence of YidC2 than in the wildtype. In a study in B subtilis using BN PAGE and C terminally His tagged and over expressed YidC homologs (SpoIIIj and YqjG) researchers found that both co purify with a ~600

PAGE 137

137 kDa complex (the F 1 F o ATP synthase) (116) In a study of human Oxa1L to evaluate its involvement in the biogenesis of the F 1 F o ATP synthase using BN PAGE combined with Western blot, researchers fou nd that Oxa1L was part of a 600 700 kDa complex containing the F 1 F o ATP synthase (181) In addition they noted that Oxa1L was also part of several lower molecular weight complexes, evidenced by a streak in the second dimension BN/SDS PAGE Western. Difference s in Membrane Protein Complexes B etween Wild Type NG8 and y idC Mutants in S. mutans Determined by BN PAGE/LC MS/MS Difference s in first dimension Blue Native p olyacrylamide gels between wild type NG8 and the yidC mutants were further a nalyzed by excision of bands followed by protein identification using LC MS/MS. The locations of the b ands that were excised from the Blue Native gels are shown in Figure 4 5. There were four gel slices analyzed for each s train Proteins that were identified by LC MS/MS are summarized in Tables 4 1 th rough 4 3. In each summary table the number of peptides recognized by mass s pectrometry and used to identify the indicated proteins from two separate BN polyacrylamide gels, are shown. Summary Table 4 1 shows the results for proteins indentified in high molecular weight complexes from Bands 1 and 2 and are referred to as the ~700 kDa and ~400 kDa complex es respectively. Shown in Table 4 2 are the proteins identified in the lower molecular weight complexes that were located in Bands 3 and 4, referred to as the ~70 kDa and ~64 kDa complexes. Table 4 3 shows a summary of the ribosomal proteins indentified for each strain in all four bands that were analyzed D etailed description s of each protein indentified in Tables 4 1 and 4 2 ( including accession numbe r, predicted molecular weight, biochemical pathway, and gene) are included in Table 4 4. I nformation about the ribosomal proteins is detailed

PAGE 138

138 (accession number, predicted molecular weight, ge ne, and additional information) in Tables 4 5 and 4 6. Tables 4 7 through 4 25, are organized according to the band that was excised and protein categories. They i ndicate the number of peptides identified by mass spec trometry with the percentage of the identified protein covered by the peptides, for both Trial 1 and Trial 2 Tables 4 26 through 4 29 show the ribosomal proteins indentified, organized by band location Trial 1 and Trial 2 indicate two separate mass spec trometry analyses of two different BN PAGE experiments using the same membrane samples. When samples were compared, a greater number of peptides corres ponding to a given protein was interpreted to mean there was more of that protein in the sample resulting in a higher percentage of protein coverage. Glycolytic enzymes There were a number of gly colytic enzymes identified, with differences in their abundanc e and location between the wild type and yidC mutants. S everal glycolytic enzymes with a greater abundance in the wildtype membranes compared to either yidC mutant, were located in high molecula r weight complexes (Tables 4 1, 4 2, 4 7 through 4 10). These included pyruvate formate lyase bifunctional acetaldehyde CoA/alcohol dehygrogense, and enolase. Additionally, there was a large increase in the number and variety of glycolytic enzymes in lo w er molecular weight complexes in both mutants with a more pronounced effect in the yidC2 mutant (Tables 4 2, 4 9 and 4 10). These proteins included glucose kinase, glucose 6 phosphate insomerase, fructose bisphosphate aldolase transketolase, glyceralde hyde 3 phosphate dehydrogenase, enolase, pyruvate kinase, and lactate dehydrogenase.

PAGE 139

139 Pyruvate formate lyase (PFL) was found in the ~70 0 kDa band from NG8 and was also found in the ~300 kDa band in all three strains, but with a larger number of peptides indentified in NG8 then either yidC mutant. Bifunctional acetaldehyde CoA/alcohol dehydrogenase (AdhE) was also found in the ~700 kDa band of NG8 but was lacking or under represented in the yidC mutants in this complex. AdhE is located downstre am of pyruvate formate lyase leading to the production of ethanol from acetyl Co A and acetaldehyde with 2 NADH oxidized to 2 NAD+, thus maintaining redox balance in the cell. In additi on to the ~700 kDa complex, AdhE was found in the ~70 and ~64 kDa bands in NG8 and yidC1 mutant, but not in the yidC2 mutant. In a t ranscriptome analysis that evaluated genes regulated by the acid tolerance response by Chen, et al 2010 (182) it was found that adhE was down r egulated in response to low pH. Thus it would be consistent for the yidC2 mutant to hav e less AdhE because of an inability to regulate cytoplasmic pH. I n a proteomics study of Clostridium thermocellum in which 2D BN/SDS PAGE ( with 1% DDM) was used, AdhE was found in a high molecular weight complex of ~800 kDa associated with membranes (183) Glucose kinase, which phosphorylates glucose tr ansported by a glucose permease, was found in both mutants in the ~ 64 kDa band but was not found in NG8 in either exper iment Phosphoglyceromutase, en olase, and pyruvate kinase were better represented in the yidC mutant samples in lowe r molecular weight complexes compared to the wild type NG8 samples, with a more pronounced effect in the yidC2 mutant strain. These enzymes are involved in the last thre e steps of glycolysis leading to the production of pyruvate. Pyruvate kinase, which functions as a homotetramer, is thought to be the rate limiting step in glycolysis in S. mutans because it is completely dependent

PAGE 140

140 on glucose 6 phosphate for activity and is inhibited by inorganic phosphate (184) Since t hese enzymes are known to function as dimers or trimers their presence of in the lower molecular weight complexe s could represent a disruption of a high molecular weight func tional multi enzyme complex in the yidC mutants. Glucose 6 phosphate isomerase, fructose bisphosphate aldolase, and transketolase, enzymes of the Pentose Phosphate Pathway and nucleotide biogenesis were, in at least one instance, found to be more highly r epresented in the y idC2 mutant than the yidC1 mutant or wild type NG8. This effect was seen primarily in the ~64 kDa complex (Tables 4 2 and 4 10). Moreover, transketolase was more highly represented in both yidC mutants in the ~64 kDa complex than in NG8 (Table 4 10). This combined with the increased level of GAPDH (glyceraldehydes 3 phosphate dehydrogenase) in the mutants, suggests that glycolytic metabolites might be diverted to the Pentose Phosphate Pathway i n the absence of Y idC1 or Y idC2 This could reflect an increased need for nucleotides due to up regulation of stress genes in the mutants. In a membrane proteomics study by Saller et al. 2010, transketolase and GAPDH of B. subtilis were found to be up re gulated two and three fold respectively when SpoIIIJ was depleted in a yqjG strain (117) Lactate deh ydrogenase converts pyruvate to lactate at the expense of NADH 2 and is located at an important branch point in metabolism. This protein was identified in all thre e strains in the ~700 kDa band in Trial 1, and only in the yidC mutants in Trial 2, ( Table 4 1 ) It was also found in equal amounts in all three strains in the ~300 kDa band ( Table 4 2 ). Lactate dehydrogena se was not represented in the ~70 kDa, but was

PAGE 141

141 found in ~64 kDa band in all three strains. In Trial 1 there was a greater representation of this protein in yidC2 than in yidC1 or NG8 (Tables 4 2 and 4 10) (NAD+) specific glyceraldehyde 3 phosphate dehydrogenase (GAPDH) was found to be highly represented in all samples submitted for LC MS/M S, but was more prominent in yidC2 mutant membranes in the lower m olecular weight complexes GAPDH ( gapC ) is an important glycolytic enzyme as it produces NADH and 1, 3 bisphosphoglycerate, which is converted to 3 phosph o glycerate with the production of one ATP by phosphoglycerate kinase It also works in reverse to produ ce glyceraldehyde 3 phosphate, which can be directed into t he Pentose Phosphate P athway for nucleotide synthesis ( S. mutans does not contain the oxidative branch of the Pentose Phosphate Pathway). Other investigators have reported changes in GAPDH amount/localization under defined circumstances. A study by Baev (185) that investigated the membrane association of a GTP binding protein induced by stress (186) found that GAPDH demonstrated an increased association with the membrane fraction during stationary phase. SGP ( Streptococcus GTP binding protein) was found to associat e with the membrane during acid stress, temperature stress and during stationary phase (185) A study by Biswas (187) that i nvest igated the effects of a deletion of the HtrA extracellular prote ase/chaperone, found that the level of extracellular GAPDH in media increased when HtrA was absent. This study also found that extracellular enolase was increased in the HtrA mutant background. Citrate metabolic e nzymes. Thre e enzymes of the Citrate C yc le were present in wild type NG8 in the ~700 kDa band, but not indentified in either the yidC 1 or yidC2 mutant s (Tables 4 1 and 4

PAGE 142

142 11 ) The three enzymes identified were a putative citrate lyas e beta subunit c itrate lyas e alpha subunit and c itrate synth ase. S. mutans has an incomple te TCA cycle, and the enzymes that it does possess are probably involved in creating precursors for amino acid synthesis (10) Ce ll wall associated p roteins A number of cell wall associated proteins were identified including; PrsA, penicillin binding protein 1a (Pbp1a), P ac/P1, and HtrA (serine protease) (Tables 4 12 to 4 14). PAc/P1 (170 kDa sortase substrate, surface adhesin) was only found in the yidC mutants in the ~700 kDa band. HtrA was found in the ~700 kDa band of the yidC2 but also in the ~70 kDa band of NG8 a nd the yidC 1 mutant, indicating there may be a difference in the localization of this serine protease in the absence of yidC2 PrsA (foldase protein with PPIase activity) was also found in the ~700 kDa band, but was present in all three strains. Additio nally, PAc/P1 was also found in the ~300 kDa band of the yidC2 mutant only. Penicilli n bindi ng protein 1a was indentified in all three strains in the ~700 kDa band, but only in the yidC mutants in t he ~ 300 kDa band. The presence of cell wall associated proteins in a high molecular weight complex associated with membranes from S. mutans may stem from an inefficiency in their secretion and subsequent accumulation and association with the membrane. The secretion of surface localized proteins is not well un derstood in streptococci but presumed to occur post translationally via the general secretio n pathway (188) It has further been proposed that secretion o ccurs through a localized micro domain known as the Exportal (68) The presence of cell wall associated proteins in the yidC mutants may indicate a defect in the assembly and/or function of the secretion machinery and Exportal. S. mutans P1

PAGE 143

143 has been reported to co localize with an Exportal like structure (189) P1 maturation is negatively affected by deletion of yidC2 as indicated by decreased immunoreactivity of certain anti P1 monoclonal antib odies with the yidC2 strain while elimination of yidC1 results in increased detection of P1 on the cell surface. The association of P1 with the membranes of the yidC mutants and not in the wildtype is consistent with the YidC involvement in th e secretion of surface localized proteins through the membrane in gram positive bacteria In addition the changes in function and antigenicity of cell surface localized P1 further suggests a role of S. mutans YidC proteins in surface protein maturation. Transport p roteins There were several transporters identified in the ~700 and ~300 kDa bands of the yidC mutants that w ere not found in NG8 (Tables 4 1, 4 15 and 4 16 ). These included an oligopeptide ABC transporter (substrate binding protein), putative mechanosensitive channel of large conductance (MscL), a putative ABC transporter (lipoprotein) and the maltose/maltodextrin ABC transporter sugar binding protein (MalX ). MscL is a known substrate of the YidC only (Sec independent) pathway in E. coli (97, 190) MalX (known as MalE in E coli ) is the substrate binding protein of the MalXFGK 2 ABC transporter, and also requires YidC for proper assembly in E. coli (99) Chaperones Several chaperone proteins were identified, with an increased representation of GroEL in the yidC1 mutant compared to NG8 or yidC2 in the higher molecular weight ~700 kD a and ~300 kDa bands (Tables 4 1, 4 17 to 4 18 ). In the lower molecular weight ~70 kDa and ~64 kDa bands (Tables 4 2, 4 19 and 4 20 ), GroEL was equally

PAGE 144

144 represented in NG8 and yidC1 but was poorly represented in yidC2 DnaK and GroES were also identified, with GroES only represented in the yidC mutants, and DnaK only found in NG8 or yidC1 Overall, there appeared to be less molecular chaperones associated with the yidC2 membranes than either NG8 or yidC1 I n the case of GroEL there was more of it associated with the yidC1 mutant in the higher molecular weight complexes Proteomics data of the yidC depletion strain in E. coli showed an increase in GroEL and DnaK in the cytoplasm and membrane under yidC depletion conditions (139) In contrast, the doubly depleted SpoIIIJ/ yqjG strain in B. subtilis did not affect the level of GroELS or DnaKJ (117) Amino acid m etabolism Proteins that were identified that are involved amino acid me tabolism are shown in Tables 4 21 to 4 23 Branched chain amino acid aminotransferase and aspartate semialdehyde dehydro genase were found in the ~64 kDa band of the yidC2 mutant, but not found in NG8 or the yidC1 mutant (Table 4 23 ) These enzymes are involved in the catabo lism and biosynthesis of branch chained amino acids. Leucine, isoleucine, and valine are found in high concentration s in membrane proteins. If membrane proteins were degraded because of an insertion defect, this pathway could be up regulated to compensate for the increase d need for catabolism of branch chained a mino acids (due to proteolysis). In add iti on to these enzymes, a branched chain amino acid ABC transporter was also identified in both yidC mutants in the ~700 kDa band but was absent in NG8 (Table 4 15 ) providing further evidence that this pathway may be up regulated or perturbed in the muta nts Furthermore, NAD+ specific glutamate dehydrogenase (GdhA) was under represented in the yidC2 mutant compared to NG8

PAGE 145

145 and the yidC1 mutant in the ~300 kDa band, ( Table 4 9 ) In a recent study by Chen, et al. 2010 (182) using microarray to investigate th e role of GlnR in acid mediated repression of genes related to glutamine and glutamate metabolism the authors mention that gdhA was downregulated by 0.58 to 0.75 fold during ATR (182) Therefore a decrease of GdhA in the yidC2 mutant could be indicative of an acid adapted response due to a decrease d ability to regulate the cytoplasmic pH. Butanoate, glutathione, and starch metabolic e nzymes Two proteins involved in butanoate metabolism were identified with increased represent ation in the yidC mutants; acetoin reductase and putative succinate semialdehyde dehydrogenase (Table s 4 24 to 4 25 ). In a proteomics study of Corynebacterium glutamicum in response to pH changes, researchers found that acetoin reductase was present in th e membrane frac tion and up regulated during low pH conditions (a t pH 6.0). Acetoin reductase ( b utA or budC in S. mutans ) results in the production of 2, 3 butanediol from pyruvate with intermediates of acetolactate and acetoin. Acetoin and 2, 3 butanedio l are neutral molecules, providing the cell with a means to eliminate pyruvate without producing acid. A study by Johansen (191) found that this pathway was induced by acetate in Enterobacter aerongens and that mutations in this pathway resulted in a lower terminal culture pH. Studies in S. mutans have also implicated acetoin reductase in acid tolerance. A 2D SDS PAGE proteomics study of soluble proteins from S. mutans grown at low pH found that acetoin reductase was significantly up regulated (192) In addition, a recent study by Chen, et al. 2010 using microarray foun d that acetoin reductase was up regulated fourfold under ATR conditions (182)

PAGE 146

146 Glutathione reduc tase is an enzyme involved in oxidative stress tolerance in S mutans with in creased activity when cells are grown with aeration (193) This enzyme was present in the ~70 kDa band in all three strains (Table 4 24 ). However, t here was a higher representation in the yidC2 mutant compared to NG8 or yidC1 Glutathione red uctase was also found in the ~6 4 kDa band in Trial 1 in both NG8 and yidC2 mutant, with a greater representation in the yidC2 mutant (Table 4 25 ). There were a number of enzymes involved in st arch m etabolism found in the ~70 kDa and ~64 kDa bands that were represented to a greater extent in the yidC mutants tha n in NG8 (Table s 4 24 and 4 25 ) with a great er e ffect in the yidC2 than the yidC1 mutant Glucose 1 phosphate adenyltransferase was identifi ed in both yidC mutants but absent in NG8. Additionally, g lycogen phosphorylase was present in all three strains in the ~ 70 kDa complex in both t rials, with a greater representation in the yidC2 mutant in Trial 1. Glycogen biosynthesis protein was also seen in the ~ 70 kDa band in the yidC1 and yidC2 mutants but was not identified in the NG8 sample. Ribosomal p roteins There was a pronounced decrease or lack of ribosomal proteins indentified in gel slices from the yidC1 and yidC2 mutants ( Tables 4 3, and 4 26 to 4 27 ). A number of ribosomal proteins were associ ated with the membranes of wild type NG8 in all bands submitted for LC MS/MS identification, with an increase in number and variety seen in the lower molecular weight complexes ( Tables 4 2 1 through 4 24 ) Conversely, t here was almost a complete lack of ribosomal proteins associated with the yidC2 mutant membranes. There was also a notable decrease seen in the yidC1 mutant, but the effect was less severe than in the yidC2 mutant Th is effect was not observed in a ffh

PAGE 147

147 mutant (data not shown), as there were a similar number of ribosomal protein s associate d with the membranes from this strain as in NG8 For a detailed description of the ri bosomal pro teins identified see Tables 4 5 and 4 6. Ribosomal proteins are small, ranging from 30 kDa to 6 kDa in size. In 2D BN/SDS PAGE analys is of these strains (Figure 4 2) there was a lack of proteins in the lower right corner of the yidC2 mutant, below 20 kDa and corresponding to the lower molecular weight complexes in the BN polyacrylamide gel In comparison in the NG8 and the yidC1 2D gels, there were lots of protein s stained in the area where ribosom al proteins are expected to run. Ribosomes are known to co purify w ith membranes, and in one study researchers found that FtsY is required for ribosomal association with the membrane in E. coli (194) However, since the FtsY protein is dispensable in S. mutans (as is the entire SRP pathwa y) YidC1 or YidC2 may play a role in ribosomal association with membranes The mass spectrometry data are consistent with a ribosome b inding function of YidC2 inv olved with co translational translocation, allowing for elimination of the SRP pathway. Cross complementation s tudies performed in Yeast indicate that this function is dependent on the C terminal tail of YidC2 (123) The current data provides the first evidence of a ribosomal association of YidC in S. mutans T wo ribosomal proteins that are not found in the UA159 genome sequence (10) were identified in NG8, L2 and L23 Using PCR, Dr. Nathan Lewis from the Brady lab has confirmed that L23 and L2 are present in NG8 and UA159 (personal communication with Nathan Lewis and L.J. Brady). This indicates there was a mistake made during the sequencing of the UA159 genome. L23 is the ribosomal protein show n to be important for Ffh and trigger factor ribosome interactions i n E. coli It is the L23 homolog Mrp20 in

PAGE 148

148 yeast mitochondria, which binds to the C terminal tail of Oxa1. These genes are also present in S. mutans strain NN2025, which was recently sequenced (195) Extracellular GAPDH A ctivity is I ncreased in yidC M utants GAPDH has been found on the surface of S. mutans a nd other s treptococci as well as a number of other cytoplasmic proteins It is not known why this occurs or how they are transported because they do not contain signal peptides. In one study of S. pneumoniae m utanolysi n digestions of cells walls were analyzed by 2D SDS PAGE combined with mass spectrometry (196) In this study, GAPDH, NADP glutamate dehydrogenase, enolase, lactate dehydrogenase, DnaK, and fructose bisphosphate aldolase, a mong other proteins, were identified as associated with the cell wall. Tryptic digestion of S pyogenes cell wall proteins combined with HPLC (high performance liquid chromatography) separation of peptides and LC MS/MS analysis, indentified 21 cytoplasmic proteins including GAPDH, enolase, 50S ribosomal proteins L7/L12/L5/L11, 30S ribosomal pro tein S8, pyruvate kinase, NADP dependent GAPDH, phospho glycerate kinase, pyruvate formate lyase, RopA (trigg er factor), and Hsp60 and Hsp10 homologs (197) The S taphyloco ccal transferrin receptor (Tpn) was identified as extracellular GAPDH, a 42 kDa protein that also possesses GAPDH activity. The functional confo rmation is a 170 kDa tetramer, which is necessary for GAPDH activity, while the monomer is capable of binding transferrin [ reviewed in (198) ]. Since there was more GAPDH associated with the membranes in the y idC2 mutant than NG8 or yidC1 (Tables 4 1 and 4 2), it was logical to test whether there was a difference in extracellular GAPDH activity. The various strains were grown overnight in THYE, and GAPDH activity of whole cells was measured (Materials and

PAGE 149

149 Methods). Consistent with the increased levels of GAPDH in the mutants seen by BN PAGE/LC MS/MS, there was a significant increas e in extracellular GAPDH activity in the mutant compared to the mutant or wild type NG8 ( Figure 4 6). The yidC2 C (AH412) strain was also evaluated for extracellular GAPDH activity, and demonstrated significantly increased surface localized GAPDH activity, at a level similar to a complete deletion of yidC2 YidC1C2 (SP13) was able to complement and restore GAPDH ac tivity toward wildtype levels in the yidC2 mutant. YidC2C1 did not complement and displayed GAPDH activity similar to that of the yidC2 mutant. When yidC1 was deleted in the yidC2 C background, there was a significant decrease in GAPDH activity compare d to yidC2 C (AH412) alone, indicatin g that whatever the cause for increased GAPDH activity, it is alleviated somewhat by deletion of yidC1 in this background. The purpose of extracellular GAPDH is unknown in S. mutans and will require further investigati on. It is possible that GAPDH is transported to the cell surface as part of a stress response, with an as yet unknown function. The increased surface localized activity seen in the mutants could be the result of an alteration in the maturation of GAPDH p romoting the formation of tetramers (the enzymatically active for m). It is possible that the cell wall is different in the mutants resulting in more GAPDH associated with the cell surface, whereas in the wildtype it is released into the culture supernatan t. Discussion Other proteomics approaches have been used to investigate the function of YidC homologs in their parent organism. A recent study by Price (139) used metabolic labeling with 15 N/ 14 N of membrane proteins combined with protein identification with

PAGE 150

150 LC MS/MS, to c ompare the effects of YidC depletion under aerobic and anaerobic growth. This study reported an increase in the number of ribosomes associated with membranes under aerobic growth conditions, as well as increased chaperones and a decrease in a number of AB C transporters. There were also e ffects seen in secreted proteins under aerobic conditions, as there was a decrease in periplasmic proteins. The effects of YidC depletion under anaerobic conditions were less severe. Protein aggregates were also isolated from YidC depleted cells in this study and found to contain a number of outer membrane proteins (indicating a secretion defect) periplasmic proteins (including DegP/HtrA and SurA /chaperone), ribosomal proteins, and several cytoplasmic chaperones. In a microarray study of the E. coli YidC depletion strain JS7131 (199) there was up regulation of genes for H flC and H flK, proteins that negatively regulate the activity of HflB/FtsH. YidC has been shown by others to interact with HflC/K and HflB/ Ft sH (139, 177) This microarray study also found YidC depletion resulted in up regulation of ribosomal genes as well as secY (authors not ed that secE and secG were not affected). Two genes that were most significantly affected by Y idC depletion were cadA and cadB which were up regulated 32.8 and 20.7 fold respectively. These genes are located in the C ad operon (CadA is a lysine decarboxylase, CadB transports cadeverine), which is turned on during external acid stress or anoxic growth conditions (200 201) CadA decarboxylates lysine producing CO 2 and cadeverine, effectively removing protons from the cytoplasm, cadeverine is then exported by CadB. The a uthors postulate that Y idC depletion resu lts in decreased cytoplasmic pH, due to impairment of the electron transport chain in the absence of YidC, inducing an acid

PAGE 151

151 stress response. S. mutans has a similar system to CadBA that is induced by low pH, known as the agmatine deiminase system (AgDS) (202) In this system agmatine enters the cell through the aguD transporter, is h ydrolysed by agmatine d eiminase ( aguA ) to N carbamolyputrescine and ammonia. N carbamolyputrescine is phosphorylated by putrescine carbamoyltransferase ( aguB ) to yield putrescine and carbamoylphosp hate. Carbamoylphosphate is in turn de phosphorylated by carbamate kinase ( aguC ) producing ATP, CO 2 and NH 3 The anti porter ( agu D ) then exchanges putrescine for agmatine (203) The AgDS was also shown to be regulated by multiple two component systems induced by acid and temperature stress (204) It would be interesting to test if AgDS is up regulated in either the yidC1 or yidC2 mutant s in future studies. The r esults from thi s BN PAGE study demonstrated differences in several bi ochemical pathways between wild type NG8 and the yidC mutants, that are consistent with finding s by Len (14) of acid adapted cells that suggest the yidC mutants may be respondi ng to low internal pH that is the result of decreased proton F 1 F o ATPase activity There was evidence in the yidC mutants that pyruvate may be diverted to alternative pathways by the increase i n Pentose P hosphate Pathway enzymes, and those involved in butanoate metabolism which results in the production of the neutral compound acetoi n. There was also an increase in or differe nce in the location of branched chain amino acid aminotransferase (ilvE) and aspartate semialdehyde dehydrogenase in the yidC2 mutant B oth are involved in branched chain amino acid synthesis or degradation S ince production of branched chain amino acids leads to production of ammonia, this would offset the increased internal pH resulting from

PAGE 152

152 decreased F 1 F o ATPase activity A number of o ther pathways were also affected by elimination of yidC1 and yidC2 includ ing enzymes of the Citrate Cycle, which were absent in high molecular weight complexes in the yidC mutants additionally, a number of transporters and cell wall associated proteins were found in high molecular weight complexes in the mutants, but were abse nt in the wildtype. There was also a decrease in the number of ribosomal proteins associated with the yidC mutant membranes in low molecular weight complexes. Overall these results suggest a role for YidC1 and YidC2 in several pathways possibly involving assembly of complexes containing glycolytic enzymes, Citrate metabolism, and tethering of ribosomes to the membrane. Furthermore, effects on ABC transporters and cell wall associated proteins and their detection in high molecular weight complexes suggest there may be a defect in protein transport resulting in a build up of these proteins in the membranes of the yidC mutants.

PAGE 153

153 Figure 4 1. Silver stained versus Coomas sie Blue G 250 stained first dimension Blue N ative polyacrylamide gels Membranes were prepared from S. mutans wild type NG8, yidC1 and y idC2 mutant strains 50 g of each sample was loaded per lane of an Invitrogen 3 12% NativePAGE Novex Bis Tris Gel Left panel show s results of gel stained with BioRad Silver Stain Plus Kit and right pane l shows results with Coomassie Blue G 250 ( see Materials and Methods for details) Arrows indicate bands of interest

PAGE 154

154 Figure 4 2 Silver stained first dimension Blue N ativ e polyacrylamide gel and second dimension Tricine SDS polyacrylamide gels Membranes were prepared from S. mutans wild type NG8, yidC1 and yidC 2 mutant strains 75 g of each sample was separated on an Invitrogen 3 12% NativePAGE Novex B is Tris Gel Gel strips were cut out, reduced (50mM DTT), alkylate d (50 mM DMA) and quenched (5 mM DTT, 20% eth anol) in LDS sample buffer Gel strips were then applied to second dimension 10% Tricine SDS polyacrylamide gels Second dimension g els were silver stained using Silver Stain Plus kit from BioRad Arrows ind icate the bands of interest, and boxes indicate the respective areas of interest in the second dimension gels.

PAGE 155

155 Figure 4 3 Western blots of Blue Native polyacrylamide gels reacted with antibodies against YidC1, YidC2 and SecY. Membranes were prepared from S. mutans wild type NG8, yidC1 and y idC2 mutant strains, and 50 g of each samp le was separated on an Invitrogen 4 16 % NativePAGE Novex Bis Tris Gel Proteins were transferred to a PDVF membrane and reacted with affinity purified antibodies against C terminal synthetic peptides (YidC1 or YidC2) or against 4 synthetic peptides of SecY (see Materials and Methods, Chapter 2 for description of antibodies) Arrow indicates a ~146 kDa band containing SecY.

PAGE 156

156 Figure 4 4 Western blot of second dimension BN/SDS polyacrylamide gels reacted with affinity purified antibodies against SecY, YidC1, or YidC2 Left panel show s representative second dimension SDS poly acrylamide silver stained gels for NG8, yidC1 and yidC2 strains Right panels show Western blots for SecY ( ~ 75 kDa, ~ 37 kDa and ~ 20 kDa bands), YidC1 (expected ~ 24 kDa band) and YidC2 (expected ~ 28 kDa band)

PAGE 157

157 Figure 4 5 Blue Native PAGE showing differences in membrane protein complex composition and indicating the gel slices analyzed by LC MS/MS, for S. mutans wildtype NG8 and mutant strains. For each sample, 5 0 proteins were loaded per lane o n a 3 12% Invitrogen Native Tris Gel (see Materials and M ethods) Lef t panel shows a representative Blue N ative gel stained with Coomassie Blue G 250 Right panel shows the other half of the same gel stained with Bio Safe Coomassie (BioR ad) with gel slices removed for LC MS/MS analysis. Bands 1 4 will be referred to in data T ables 4 1 through 4 29 summarizing proteins identified by LC MS/MS.

PAGE 158

158 Table 4 1 Summary of proteins ident ified by LC MS/MS in BN PAGE high molecula r weight co mplex es from Band s 1 and 2 Band 1 (~700 kDa) Band 2 (~300 kDa) Identified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Glycolysis Glyceraldehyde 3 phosphate dehydrogenase 4 a 0 0 7 2 7 2 3 1 5 1 6 Phosphoglyceromutase 1 0 1 0 Enolase phosphopyruvate hydratase 18 0 1 2 1 2 27 17 17 8 11 2 Pyruvate kinase 0 6 1 2 1 1 2 2 Pyruvate formate lyase 1 0 9 1 2 2 1 0 Bifunctional a cetaldehyde CoA/alcohol dehydrogenase 12 0 L(+) lactate dehydrogenase 3 0 2 3 3 1 1 1 3 1 2 1 Citrate cycle (TCA ) Putative citrate lyase CilB, citryl CoA lyase, beta subunit 4 0 Citrate lyase, alpha subunit 3 5 Citrate synthase 3 0 Cell w al l biosynthetic proteins Foldase protein PrsA 3 0 1 3 2 1 0 1 Penicillin binding prote in 1a 1 0 5 3 3 0 1 0 1 0 PAc/P1 Major cell surface adhesin 2 5 0 5 Serine protease HtrA 2 0 Transporters Branched chain amino acid ABC transporter (binding protein) 3 2 4 0 Oligopeptide ABC transporter, substrate binding protein OppA 2 0 MscL putative large conductance mechanosensitive channel 2 0 1 0 1 0 Putative ABC transporter (lipoprotein) 4 3 2 0 0 1 1 0 Maltose/maltodextrin ABC transporter, sugar binding protein 1 0 1 0

PAGE 159

159 Table 4 1 C ontinued Band 1 (~700 kDa) Band 2 (~300 kDa) Indentified proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Charperones GroEL 5 0 2 0 8 0 1 0 DnaK 1 0 GroES 0 2 0 1 Amino Acid metabolism Aspartate semialdehyde dehygrogenase 1 0 G lutamate dehydrogenase 17 11 7 15 3 0 P utative aminopeptidase P 1 0 1 0 4 0 a Number of unique peptides in the gel slice identified by LC MS/MS that we re used to identify the protein. Peptides indentified in Trial s 1 and 2 are separated by dashes. indicates no peptides were detected in either experiment.

PAGE 160

160 Table 4 2. Summary of proteins ident ified by LC MS/MS from BN PAGE low er molecul ar weight complex from Band s 3 and 4. Band 3 (~70 kDa) Band 4 (64 kDa) Identified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Glycolosis Phosphoenolpyruvate: sugar PTS enzyme 1 1 0 a 2 0 3 0 Glucose kinase 3 0 4 7 16 7 G lucose 6 phosph ate isomerase 2 1 1 0 15 4 F ructose bisphosphate aldolase 1 0 1 0 2 0 7 0 T ransketolase 4 0 5 2 5 1 6 1 5 8 17 6 G lyceraldehyde 3 phosphate dehydrogenase 1 0 4 0 11 0 4 3 0 5 13 7 P hosphoglycerate kinase (pgk) 1 0 1 0 1 1 0 4 5 2 Phosphoglyceromutase 1 0 1 0 3 1 4 2 Enolase 0 4 4 2 12 2 0 3 4 2 7 3 P yruvate kinase 3 1 5 9 14 7 4 5 5 14 11 8 Bifunctional acetaldehyde CoA/alcohol dehydrogenase 1 0 1 0 1 0 1 0 L(+) Lactate dehydrogenase 2 2 2 1 4 0 2 2 2 2 7 1 Charperones GroEL 7 6 7 8 0 1 8 5 4 10 6 0 DnaK 2 0 2 0 GroES 0 2 0 1 Amino acid metabolism Branched chain amino acid aminotransferase 4 0 Aspartate semialdehyde dehydrogenase 4 0 Butanoate metabolism A cetoin reductase 5 0 1 0 4 0 Putative succinate semialdehyde dehydrogenase 4 10 14 10 36 15 2 0 6 1

PAGE 161

161 Table 4 2. Continued Band 3 (~70 kDa) Band 4 (64 kDa) Identified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Glutathione metabolism A minopeptidase N, PepN 1 0 5 0 Gl utathione reductase 4 1 3 2 13 4 1 0 3 0 Starch and Sucrose metabolism G lucose 1 phosphate adenylyltransferase 4 1 1 0 9 4 P utative glycogen biosynthesis protein GlgD 1 0 5 0 G lycogen phosphorylase phsG 3 4 4 5 14 3 a Number of unique peptides in the gel slice identified by LC MS/MS that were used to identify the protein. Peptides indentified in Trial s 1 and 2 are separated by dashes. Indicates no peptides were detected in either experiment.

PAGE 162

162 Table 4 3. Summary of ribosomal proteins identified by LC MS/MS from BN PAGE experiments. Band 1 (~700 kDa) Band 2 (~300 kDa) Band 3 (~70 kDa) Band 4 (64 kDa) Protei n NG8 yidC1 yidC2 NG8 yidC1 yidC2 NG8 yidC1 yidC2 NG8 yidC1 yidC2 50S L1 1 0 2 0 3 0 50S L2 2 0 a 2 0 1 0 1 0 1 0 2 0 50S L5 0 1 0 1 0 3 0 2 50S L6 2 1 1 0 2 3 1 0 50S L10 1 0 1 0 2 0 50S L14 2 0 50S L17 1 0 1 0 1 0 1 0 1 0 1 0 2 0 50S L18 3 0 1 0 2 0 2 0 1 0 50S L19 3 0 1 50S L20 1 0 1 0 2 0 50S L22 2 3 1 3 50S L23 0 3 0 1 0 3 50S L27 0 1 0 2 30S S2 3 0 1 0 30S S3 2 0 30S S4 1 0 0 2 2 5 1 4 30S S5 7 8 2 0 1 0 0 3 0 1 0 4 30S S7 1 4 0 1 3 4 0 1 30S S8 3 2 1 1 2 3 0 2 2 0 30S S9 2 0 1 0 1 0 1 0 30S S10 0 2 0 2 30S S11 1 0 2 0 30S S12 3 0 2 0 a Number of unique peptides in the gel slice identified by LC MS/MS that were used to ID the protein. Number of p eptides indentified in Trial s 1 and 2 are separated by dashes. Indicates no peptides were detected in either experiment.

PAGE 163

163 Table 4 4 Information regarding proteins identified by LC M S/MS listed in Tables 4 1 to 4 2 including predicted location, molecular weight, gene name, biochemical pathway and accession number. Accession number Protein ID Biochemical Pathway Gene M.W. (Da) PSORTb v. 3.0.2 Glycolysis Central Metabolism 24370669 Enolase phosopyruvate hydratase E.C. 4.2.1.11 Smu.1247 eno 46,727 Cyto (10.0) a 24379618 Pyruvate kinase E.C.2.7.1.40 Smu.1190 pykF 54,236 Cyto (7.50) 24378857 Glyceraldehyde 3 P dehydrogenase E.C.1.2.1.12 Smu.360 gapC 35,937 Cyto (9.97) 24378663 Bifunctional a cetaldehyde CoA/alcohol DH E.C.1.1.1.1 Smu.148 adhE 96,862 Cyto (9.97) 24379547 L(+) lactate dehydrogenase E.C.1.1.1.27 Smu.1115 ldh 35,114 Cyto (9.97) 24379148 Phosphoenolpyruvate sugar PTS enzyme I PTS Smu.675 ptsA 63,267 Cyto (10.0) 24378895 Pyruvate formate lyase E.C.2.3.1.54 Smu.402 pfl 87,475 Cyto (9.97) 24379074 Phosphoglyceromutase E.C.5.4.2.1 Smu.596 gpmA 25,919 Cyto (7.50) 24379024 Putative glucose kinase E.C.2.7.1.2 Smu.542 glk 33,504 Cyto (9.97) 24378809 Glucose 6 phosphate isomerase E.C.5.3.1.9 Smu.307 pgi 49,291 Cyto (9.67) 24378620 Fructose bisphosphate aldolase E.C.4.1.2.13 Smu.99 fbaA 31,295 Cyto (7.50) 24378858 Phosphoglycerate kinase E.C.2.7.2.3 Smu.361 pgk 41,914 Cyto (9.97) 24378794 Transketolase E.C.2.2.1.1 Smu.291 tkt 70,945 Cyto (7.50) Citrate Cycle (TCA) 24379459 Putative citrate lyase, citryl subunit E.C.4.1.3.6. Smu.1020 cilB 32,808 Cyto (9.97) 24379460 Citrate Lyase subunit E.C.4.1.3.6. Smu.1021 cilA 55,324 Cyto (9.97) 24379144 Citrate synthase E.C.2.3.3.1. Smu.671 citZ 42,580 Cyto (9.97) Cell wall biosynthetic proteins 24379121 PrsA foldase protein (PPIase) E.C.5.2.1.8 Smu.648 prsA 39,961 Memb (9.68) + 24378955 Penicilin binding protein 1a NA Smu. 467 pbp1A 77,631 Memb (8.77) 24379087 PAc surface adhesin P1 NA Smu.610 spaP 169,841 C Wall (10.0) + 24380491 HtrA serine protease (DegP) NA Smu.2164 htrA 42,941 Unknown Transport 24378763 Oligopeptide AB C transporter subs binding protein NA Smu.255 oppA 60,100 C Wall (9.21) 24380047 Branched chain amino acid ABC transporter NA Smu.1669 livK 41,087 Non Cyto c 24379274 Putative MscL NA Smu.819 mscL 13,494 Memb (10.0) 24379952 Maltose/maltodextrin ABC transporter NA Smu.1568 malX 45,043 Unknown 24379553 Putative ABC transporter (Lipoprotein) NA Smu.1121c yufN 36,510 Non Cyto

PAGE 164

164 Table 4 4 Continued Accession Number Protein ID Biochemical Pathway Gene M.W. (Da) PSORTb v. 3.0.2 Chaperones 24380300 GroEL 60 kDa chaperonin NA Smu.1954 groEL 56,970 Cyto (9.97) 24380301 GroES 10 kDa chaperonin NA Smu.1955 groES 9,907 Cyto (9.97) 24378606 DnaK heat shock protein 70 NA Smu.82 dnaK 65,155 Cyto (9.97) Amino Acid Metabolism 24380211 Aminopeptidase P E.C. 3.4.11.9 Smu.1850 pepP 39,568 Cyto (9.97) 24379429 Aspartate semialdehyce dehydrogenase E.C.1.2.1.11 Smu.989 asd 38,772 Cyto (9.95) 24379360 Glutamate dehydrogenase, NAD+ specific E.C.1.4.1.4 Smu.913 gdhA 48,102 Cyto (9.67) Butanoate Metabolism 24379736 Acetoin reductase putative acetoin dehydrogenase E.C.1.1.1.5. Smu.1322 budC 26,773 Cyto (9.67) 24380458 Putative succinate semialdehyde dehydrogenase E.C.1.2.1.16 Smu.2127 50,328 Cyto (9.67) Glutathione Metabolism 24379293 Glutathione reductase E.C.1.8.1.7 Smu.838 gshR 48,847 Cyto (9.67) 24379563 Aminopeptidase N E.C.3.4.11.15 Smu.1132 pepN 96,701 Cyto (9.97) Starch Metabolism 24379927 Glucose 1 phosphate adenyltransferase E.C. 2.7.7.27 Smu.1538 glgC 42,069 Cyto (7.50) 24379924 Glycogen phosphorylase E.C.2.4.1.1 Smu.1535 phsG 90,920 Cyto (7.50) 24379926 Glycogen biosynthesis protein E.C.2.7.7.27 Smu.1537 glg D 42,204 Cyto (7.50) a Location prediction is based on PSORTb v3.0.2 ( http://www.psort.org/psortb/ ) S cores closer to 10 have a better chance of being located in the predicted location. b Unknown, protein had a 2.50 score for cytoplasm, cytoplasmic membrane, cell wall, and extracellular c Protein had a signal peptide dete cted and a 3.33 score for extra cellular, 3.33 cell wall, and 3.33 for cytoplasmic membrane so protein is not located in cytoplasm + Protein contains a signal sequence

PAGE 165

165 Table 4 5. Details of 50S ribosomal proteins in dentified by LC MS/MS inc luding accession number, molecular weight, gene, and additional information. Accession N umber Ribosom al P roteins Gene M W ( Da ) Location PSORTb Additional Information 24380006 50S L1 Smu.1626 rplA 24 379 Cyto (9.97) F orms part of the L1 stalk along with 23S rRNA Also f unctions as a translation repressor that binds its own mRNA. 157150843 50S L2 Sgo.1982 a rplB 29,794 Cyto (7.50) No t found in S. mutans UA159 sequence Is required for association of the 30S and 50S subunits to form the 70S ribosome 2438035 50S L5 Smu.2015 rplE 19,680 Cyto (7.50) P art of 50S and 5S/L5/L18/L25 subcomplex; contacts 5S rRNA and P site tRNA; forms a bridge to the 30S subunit in the ribosome by binding to S13 24380354 50S L6 Smu.2011 rplF 19,305 Cyto (7.50) M utations confer resistance to aminoglycoside antibiotics such as gentamicin localized to C terminal domain 24379400 50S L10 Smu.957 rplJ 17,536 Cyto (9.97) B inds the two ribosomal protein L7/L12 dimers and anchors them to the large ribosomal subunit 24380359 50S L14 SMU.2017 rplN 13 kDa Cyto (9.67) B inds to the 23S rRNA between the centers for peptidyl transferase and GTPase 24380342 50S L17 Smu.2000 rplQ 14,409 Cyto (9.97) A component of the macrolide binding site in the peptidyl transferase center 24380353 50S L18 Smu.2010 rplR 12,763 Cyto (9.67) B inds 5S rRNA along with protein L5 and L25 24379704 50S L19 Smu.1288 rplS 13,021 Cyto (9.97) L ocated at the 30S 50S ribosomal subunit interface 24379170 50S L20 Smu.699 rplT 13,620 Cyto (9.67) B inds directly to 23S ribosomal RNA prior to in vitro assembly of the 50S ribosomal subunit 24380364 50S L22 Smu.2022 rplV 12,313 Cyto (9.67) B inds to 23S rRNA during 50S assembly ; makes contact with all 6 domains of the 23S rRNA in the assembled ribosome. M utations in this gene resu lt in erythromycin resistance. 290581287 50S L23 Smu 2025.1761 rplW 10,855 Unknown Not listed in S. mutans UA159 but is in S. mutans NN2025. Part of exit tunnel. 24379304 50S L27 SMU.849 rpmA 10,290 Cyto (9.67) Involved in the peptidyltransferase reaction during translation 24380351 50S L30 SMU.2008 rpmD 6,250 Cyto (9.67) L30 binds domain II of the 23S rRNA and the 5S rRNA; similar to eukaryotic protein L7 a Peptide matched to a protein from S mutans NN2025 Streptococcus mutans UA159 sequence does not list this protein.

PAGE 166

166 Table 4 6. Details of 30S ribosomal proteins i dentified by LC MS/MS including accession number, molecular weight, gene, and additional information. Accession N umber Ribosom al P rotein s Gene M W ( Da ) Location PSORTb Additional Information 24380374 30S S2 Smu.2032 rpsB 28,962 Cyto (9.97) O ne of the last subunits in t he assembly of the 30S subunit and is required for an active 30S subunit 161486819 30S S3 Smu.2021 rpsC 23,991 Cyto (9.97) F orms a complex with S10 and S14, binding to the lower part of the 30S subunit head and the mRNA in the complete ribosome 24380465 30S S4 Smu.2135c rpsD 22,903 Cyto (7.50) P rimary rRNA binding protein, nucleates 30S assembly and is involved in translational accuracy with proteins S5 and S12 24380352 30S S5 Smu.2009 rpsE 16,990 Cyto (9.97) Located at back of 30S subunit, plays a role in translational accuracy. Mutations result in spectinomycin resistance 24378855 30S S7 SMU.358 17,674 Cyto (7.50) B inds directly to 16S rRNA where it nucleates assembly of the head domain of the 30S subunit 24380355 30S S8 Smu.2012 rpsH 14,562 Cyto (9.97) B inds directly to 16S rRNA central domain where it helps coordinate assembly of the platform of the 30S subunit 24378685 30S S9 Smu.170 rpsI 14,089 Cyto (9.67) Is in direct contact with the tRNA during translation 24379400 30S S10 Smu.957 rpsJ 17,536 Cyto (9.97) Binds two ribosomal protein L7/L12 dimers and anchors them to the large ribosomal subunit 24380344 30S S11 Smu.2002 rs11 13,268 Cyto (9.67) Located on the 30S subunit platform, bridges several RNA helices of the 16S rRNA; forms part of the Shine Dalgarno cleft in the 70S ribosome; interacts with S7 and S18 and IF 3 24378854 30S S12 Smu.357 rps L 14,978 Cyto (9.67) I nteracts with bases of 16S rRNA i nvolved in tRNA selection in A site and with mRNA backbone. Located at interf ace of the 30S and 50S subunits. Mutations confer streptomycin resistance

PAGE 167

167 Table 4 7 Glycolytic enzymes ident ified by LC MS/MS from BN PAGE B and 1 ( ~700 kDa complex ) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC 2 NG8 yidC 1 yidC2 Indentified Proteins Trial 1 C Trial 1 Trial 1 Tria l 2 Tri al 2 Trial 2 Glycerald ehyde 3 phosphate dehydrogenase 4 a 12% b 2 7% d 7, 24% 7 22% Phosphoglyceromutase 1, 11% Enolase 18, 59% 1, 4% 1 3% 2, 6% 2 6% Pyruvate kinase 1 2% 6 15% 2 4% Pyruvate formate lyase 1, 1% L(+) lactate dehydrogenase 3, 11% 2, 11% 3 15% 3 12% 1 4% Bifunctional a cetald ehyde CoA/alcohol dehydrogenase 12, 17% a Number of unique peptides in the gel slice identified by LC MS/MS t hat were used to ID the protein b P ercent of indentified protein covered by the peptides c Gel slices from two separate gels were submitted for LC MS/MS analysis and are rep resented as Trial 1 and T rial 2 d indicates not detected. Table 4 8 Glycolytic enzymes identified by LC MS/ MS from BN PAGE B and 2 ( ~300 kDa complex ) including the number of peptides and percentage of protein represented for each Trial NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Glyceraldehyde 3 phosphate dehydrogenase 2 7% 1 5% 1 3% 3 12% 5, 17% 6, 20% Phosphoglyceromutase 1 11% Enolase 27 65% 17 63% 11 43% 17 42% 8 23% 2 6% Pyruvate kinase 1 5% 2 7% 1 2% 2 4% Pyruvate formate lyase 9 14% 2 5% 1 3% 1 1% 2 3% L(+) lactate dehydrogenase 1 4% 3 16% 2 11% 1 4% 1 4% 1 4% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total protein represented next to it indicates not detected.

PAGE 168

168 Table 4 9 Glycolytic enzymes ident ified by LC MS/MS from BN PAGE B and 3 ( ~70 kDa complex ) includin g the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Phosphoenolpyruvate: sugar PTS enzyme I 1 3% 2 7% 3 5% Putative glucose kinase 3 15% Glucose 6 phosphate isomerase 2, 4% 1 3% Fructose bisphosphate aldolase 1 8% Transketolase 4 10% 5 16% 5 10% 2 4% 1 3% Glyceraldehyde 3 phosphate dehydrogenase 1 5% 4 22% 11 48% Phosphoglycerate kinase 1 5% 1 4% Phosphoglyceromutase 1 11% Enolase 4 16% 12 46% 4 14% 2 6% 2 6% Pyruvate kinase 3 10% 5 15% 14 38% 1 3% 9 18% 7 16% Bifunctional a cetaldehyde CoA/alcohol dehydrogenase 1 3% 1 3% L(+) lactate dehydrogenase 2 11% 2 11% 4 14% 2 7% 1 4% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total protein represented next to it indicates not detected. PTS phosphotransferase system.

PAGE 169

169 Table 4 10 Glycolytic enzymes identified by LC MS/MS from BN PAGE B and 4 ( ~64 kDa complex ) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial2 Trial 2 Putative glucose kinase 4 15% 16, 53% 7, 25% 7, 20% Glucose 6 phosphate isomerase 1, 6% 15 46% 4, 17% Fructose bisphosphate aldolase 1 5% 2 13% 7, 35% Transketolase 6 20% 5, 17% 18, 40% 1, 2% 8, 14% 6, 11% Phosphoglycerate kinase 1 5% 5, 15% 1, 4% 4, 11% 2, 8% Phosphoglyceromutase 1 1 11% 3, 20% 4, 27% 1, 4% 2, 9% Enolase 4, 16% 7, 26% 3, 11% 2, 6% 3, 9% Pyruvate kinase 4 12% 5, 14% 11, 29% 5, 14% 14, 31% 8, 20% Glyceraldehyde 3 phosphate dehydrogenase 4 17% 13, 60% 3, 13% 5, 22% 7, 26% Bifunctional a cetaldehyde CoA/alcohol dehydrogenase 1 2% 1, 3% L(+) lactate dehydrogenase 2 11% 2, 11% 7, 28% 2, 7% 2, 7% 1, 4% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated in parentheses, with the percent of the total protein represented next to it indicates not detected.

PAGE 170

170 Table 4 11 Proteins identified by LC MS/MS in BN PAGE gel B and 1 (~700 kDa) involved in c itrate metabolism including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Putative citrate lyase CilB, citryl CoA lyase, beta subunit 4 20% Citrate Lyase, alpha subunit 3, 7% 5, 10% Citrate synthase 3, 8% As in Table 4 7 the number of unique peptides identified by LC M S/MS is indicated with the percent of the total protein represented next to it indicates not detected

PAGE 171

171 Table 4 12 Cell wall associated proteins identified by LC MS/MS in BN PAGE gel B and 1 (~700 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 PrsA foldase protein 3, 9% 1, 6% 2, 11% 3, 11% 1, 5% Penicillin binding protein 1a 1, 2% 5, 12% 3, 6% 3, 7% PAc/P1 surface adhesin 1, 1% 2, 2% 5 4% HtrA serine protease 2, 6% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total protein represented next to it. indicates not detected Table 4 13 Cell wall associated proteins iden tified by LC MS/MS in BN PAGE gel B and 2 (~300 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 PrsA foldase protein 1, 5% Penicillin binding protein 1a 1 2% 1 2% PAc/P1 surface adhesin 5, 4% Table 4 14 Cell wall associated proteins iden tified by LC MS/MS in BN PAGE gel B and 3 (~70 kDa) including the number of peptides and percentage of protein represented for each Trial NG8 yidC1 yidC 2 NG8 yidC1 yidC 2 I dentified Proteins Trial 1 Trial 1 Trial1 Trial 2 Trial 2 Trial 2 HtrA 2, 10% 2, 11%

PAGE 172

172 Table 4 15 Proteins ide ntified by LC MS/MS in BN PAGE B and 1 (~700 kDa) involved with transport including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Branched chain amino acid ABC transporter 3 8% 4, 12% 2, 6% Oligopeptide ABC transporter, substrate binding protein OppA 2, 5% Putative large conductance mechanosensitive channel 2, 17% 1, 17% Putative ABC transporter (lipoprotein) 4, 21% 2, 7% 3, 10% Maltose/maltodextrin ABC transporter, sugar binding protein 1, 4% 1, 4% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total protein represented next to it. indicates not detected Table 4 16 Proteins identi fied by LC MS/MS in BN PAGE B and 2 (~300 kDa) involved with transport including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Putative large conductance mechanosensitive channel 1, 17% Putative ABC transporter (lipoprotein) 1, 9%

PAGE 173

173 Table 4 17 Chaperone proteins iden tified by LC MS/MS in BN PAGE gel B and 1 (~700 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 GroEL 60 kDa chaperonin 5, 20% DnaK heat shock protein 70 1 4% GroES 10 kDa chaperonin 2 21% 1 12% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total pr otein represented next to it. indicates not detected Table 4 18 Chaperone proteins iden tified by LC MS/MS in BN PAGE gel B and 2 (~300 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 GroEL 60 kDa chaperonin 2, 4% 8, 24% 1, 2% Table 4 19 Chaperone proteins iden tified by LC MS/MS in BN PAGE gel B and 3 (~70 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 GroEL 60 kDa chaperonin 7, 24% 7, 24% 6, 15% 8, 17% 1, 3% DnaK heat shock protein 70 2, 7% 2, 7% GroES 10 kDa chaperonin 2, 21% 1, 12% Table 4 20 Chaperone proteins iden tified by LC MS/MS in BN PAGE gel B and 4 (~64 kDa) including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Protein s Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 GroEL 60 kDa chaperonin 8, 25% 4, 14% 6, 16% 5, 11% 10, 22%

PAGE 174

174 Table 4 21 Proteins identif ied by LC MS/MS in BN PAGE gel B and 1 (~700 kDa) involved in amino acid metabolism including the number of peptides and percentage of protein represented for each Trial NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Aspartate semialdehyd e dehydrogenase 1, 6% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indicated, with the percent of the total protein represented next to it. indicates not detected Table 4 22 Proteins identif ied by LC MS/MS in BN PAGE gel B and 2 (~300 kDa) involved with amino acid metabolism including the number of peptides and percentage of protein represented for each Trial. NG8 yidC1 yidC2 NG8 yidC1 yidC2 Identified Protein s Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Aminopeptidase P 1, 3% 1, 3% 4, 16% Glutamate dehydrogenase, NAD+ specific 17, 47% 7, 27% 3, 10% 11 28% 15 43% Table 4 23 Proteins identif ied by LC MS/MS in BN PAGE gel B and 4 (~64 kDa) involved with amino acid metabolism including the number of peptides and percentage of protein represented for each Trial NG8 yidC1 yidC2 NG8 yidC1 yidC2 I dentified Protein s Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Branched chain amino a cid amino transferase 4, 17% Aspartate semialdehy e dehydrogenase 4, 17%

PAGE 175

175 Table 4 24 Proteins identif ied by LC MS/MS in BN PAGE gel B and 3 (~70 kDa) involved in butanoate, glutathione and starch met abolism including the number of peptides and percentage of protein represented for each Trial. NG 8 yidC1 yidC2 NG 8 yidC1 yidC2 Identified Proteins Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Butanoate m etabolism Acetoin reductase putative acetoin dehydrogenase 5, 29% Putative succinate semialdehyde dehydrogenase 4, 16% 14, 47% 36, 82% 10, 21% 10, 22% 15, 32% Glutathione M etabolism Glutathione reductase 4, 20% 3, 12% 13 49% 1, 2% 2, 5% 4, 14% Amino peptidase N 1, 2% 5, 8% Starch Metabolism Glucose 1 phosphate adenyltransferase 4, 14% 1, 3% Glycogen phosphorylase 3, 7% 4, 10% 14, 22% 4, 5% 5, 12% 3, 7% Glycogen biosynthesis protein 1, 6% 5, 16% As in Table 4 7 the number of unique peptides identified by LC MS/MS is indi cated with the percent of the total protein represented next to it indicates not detected Table 4 25 Proteins identif ied by LC MS/MS in BN PAGE gel B and 4 (~6 4 kDa) involved in butanoate, glutathione and starch metabolism including the number of peptides and percentage of protein represented for each Trial. NG 8 yidC1 yidC2 NG 8 yidC1 yidC2 I dentified Protei ns Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 Butanoate m etabolism Acetoin reductase putative acetoin dehydrogenase 1, 11% 4, 30% Putative succinate semialdehyde dehydrogenase 2, 9% 6, 25% 1, 3% Glutathione M etabolism Glutathione reductase 1, 8% 3, 21% Starch m etabolism Glucose 1 phosphate adenyltransferase 1, 7% 9, 33% 4, 13%

PAGE 176

17 6 Table 4 26 Ribosomal proteins identified by LC MS/MS in BN PAGE g el B and 1 (~700 kDa) including the number of peptides and percentage of protein represented for each Trial. I ndentified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 50S L2 a 2, 11% 50S L5 1, 10% 50S L17 1, 14% 50S L18 3 43% 30S S2 3 13% 30S S3 2 13% 30S S4 1, 11% 30S S9 2 15% a Peptides matched to a protein from S mutans NN2025 ( S. mutans UA159 genomic sequence does not list 50S L 2). BN PAGE experiments were per formed on S. mutans strain NG8. Ta ble 4 27 Ribosomal proteins identified by LC MS/MS in BN PAGE gel B and 2 (~300 kDa) including the number of peptides and percentage of protein represented for each Trial. Proteins I ndentified NG8 yidC1 yidC2 NG8 yidC1 yidC2 Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 50S L2 a 2, 12% 1, 5% 50S L5 1, 10% 50S L14 2, 19% 50S L17 1, 14% 50S L18 2, 25% 50S L19 3, 28% 2 20% 30S S4 2 10% 30S S5 7, 54% 2, 12% 1, 7% 8 59% 30S S9 1, 8% a Peptides matched to a protein from S mutans NN2025 ( S. mutans UA159 genomic sequence does not list 50S L 2). BN PAGE experiments were per formed on S. mutans strain NG8.

PAGE 177

177 Table 4 28 Ribosomal proteins identified by LC MS/MS in BN PAGE gel B and 3 (~70 kDa) including the number of peptides and percentage of protein represented for each Trial. I dentified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 50S L1 1 7% 50S L2 a 1 6% 1 5% 50S L5 3, 22% 50S L6 2, 15% 1 16% 1 11% 50S L10 1 17% 1 17% 50S L17 1 14% 1 14% 50S L18 2 25% 50S L19 1 10% 50S L20 1, 18% 1, 18% 50S L22 2 25% 3 35% 50S L23 b 3 22% 1 10% 50S L27 1 15% 30S S4 2 10% 5 19% 30S S5 3 25% 1 7% 30S S7 1 9% 4 28% 1 9% 30S S8 3 39% 1 14% 2 14% 1 14% 30S S10 2, 25% 30S S11 1 11% 30S S12 3 24% a Peptides matched to a protein from S mutans NN2025 ( S. mutans UA159 genomic sequence does not list 50S L2). BN PAGE experiments were per formed on S mutans strain NG8. b Peptides matched to L23 from S. mutans NN2025, S. mutans UA159 genomic sequence does not list 50S L23 and the genome sequence is no t available for NG8, the strain used in these experiments.

PAGE 178

178 Table 4 29 Ribosomal proteins identified by LC MS/MS in BN PAGE gel B and 4 (~64 kDa) including the number of peptides and per centage of protein represented for each Trial. I dentified Proteins NG8 yidC1 yidC2 NG8 yidC1 yidC2 Trial 1 Trial 1 Trial 1 Trial 2 Trial 2 Trial 2 50S L1 2 11% 3 20% 50S L2 a 2 11% 50S L5 2, 16% 50S L6 2 15% 1, 16% 3, 26% 50S L10 2, 13% 50S L17 1 14% 1 14% 2 20% 50S L18 2 25% 1 25% 50S L20 2 28% 50S L22 1, 10% 3 26% 50S L23 b 3 22% 50S L27 2 25% 50S L30 2 50% 1 23% 30S S2 1 4% 30S S4 1, 5% 4, 16% 30S S5 4 35% 30S S7 3 35% 4 29% 1, 9% 30S S8 2 29% 2, 14% 3 35% 2 14% 30S S9 1 8 % 1 8% 30S S10 2 25% 30S S11 2 24% 30S S12 2 9% a Peptides matched to a protein from S mutans NN2025 ( S. mutans UA159 does not have 50S L2). BN PAGE experiments were per formed on S. mutans strain NG8. b Peptides matched to L23 from S. mutans NN2025, S. mutans UA159 does not have 50S L23 and the genome sequence is not available for NG8, the st rain used in these experiments

PAGE 179

179 Figure 4 6. Extracell ular GAPDH activity in S. mutans NG8 wild type and various y idC mutants Results are expressed as OD 340 standardized to the OD 600 of the cells used for the assay The a ssay was completed in triplicate with overnight cultures of each strain grown in THYE. The s tan dard deviations of the average are indicated by error bars. Statistically significant differences are indicated (* indicates P 0.02, ** indicates P 0.001 ) and were test.

PAGE 180

180 CHAPTER 5 CONCLUSIONS AND FUTURE DIRECTIONS Development of Tools to Examine Compensatory or Redundant Functions in Membrane Biogenesis Using E. coli as a model organism, techniques have been developed to purify membrane proteins and reconstitute them into proteoliposomes. This combined with in vitro transcr iption and translation has allowed many advances to be made in the field of protein secretion and transloc ation thus facilitating d issection of the minimum requirements for this vital process. Additionally, the implementation of conditional expression systems to evaluate essential proteins in these pathways has furthered our knowledge of the in vivo requireme nts. Most of these studies have inv olved E. coli with a few examples in Bacillus subtilis Recent discoveries in gram positive streptococcal species such as ; the non essentiality of the SRP pathway in S. mutans and S. pyogenes ; the possibility of an Exportal micro domain for secreted proteins; the existence of accessory SecA2/SecY2 in S. gordonii S. pneumoniae and S. parasanguinis ; and the presence of two Yid C proteins with different functions in S. mutans indicate there are l ikely other major differe nces between what is known in E coli and what remains to be learned in the s treptococci. T here are still signif icant gaps in our knowledge of s treptococcal protein translocation and membrane biogenesis that need to be filled befo re we can reach the same level of understanding available for E. coli and B. subtilis. Streptococci have several redundant pathways for protein translocation and secretion, making it difficult to determine and/or evaluate essential functions of proteins in secretion For example there are two YidC homologs, including one that may cooperate with the SRP to enable efficient co translational translocation. Through work

PAGE 181

181 conducted in this study a number of tools have been developed, which can be used in future experiments to dissect the pathways involved in membrane biogenesis and secretion A conditional expression system was developed using the PcelB promoter to allow controlled expression of yidC2 while eliminating yidC1. Depl etion conditions were e valuated for this system and it was found that growth in TDM 0.5% mannose for 5 to 7 hours resulted in depletion of Y idC2 Further optimization with different combinations of sugars may improve expression control, and improve the speed at which carbon cat abolite repression occurs, resulting in more efficient repression It should also be possible to use this system to conditionally express other essential genes for example generation of strain s that are depleted in YidC2 and the SRP pathway and/or compo nents of the SecYEG translocon The affinity purified antibodies described in Chapter 2 against C terminal peptides of YidC1 and YidC2 and the cytoplasmic loop between TMD 2 and 3 of YidC2 will be useful reagents in future experiments. These antibodies can be covalently coupled to agarose beads and used in immunoprecipitation experiments to determi ne protein protein interactions with the YidC proteins of S. mutans Additionally a peptide antibody against SecY was also created and affinity purified to pr oduce a mono specific reagent that can be used to track SecY in S. mutans This antibody may also be used in immun oprecipitation experiments to i dentify reacting partners of the u ncharacterized SecY translocon of S. mutans The development, purification, and characterization of these reagents against S. mutans membrane translocation components are necessary tools that can be used in combination with other previously made antibodies against S.

PAGE 182

182 Ffh and FtsY to continue to d ecipher the proteins involved in targeting and insertion into S. mutans membranes. A Function of the C Terminal Tail of YidC2 in Stress Tolerance A number of strains were constructed to investigate functional difference s of the two YidC proteins in S. mut ans These strains were evaluated by grow th curve under non stress, acid stress and osmotic stress conditions. R esults indicated a clear role of the C terminal tail of YidC2 in stress toleran ce. When the C terminal tail was deleted an intermediate grow th phenotype was see n under non stress conditions. H owever when exposed to acid or osmotic stress this partial mutant strain grew similar ly to a complete deletion of yidC2. When yidC1 was deleted in the y idC2 C background a worse grow th phenotype than complete elimination of yidC2 resulted C himeric YidC1 C2 ( with the C termi nal tail of yidC2 ) restored stress tolerance to the yidC2 mutant, while YidC2C1, caused a d ominant negative effect with slower growth than the un complemented yidC2 mutant indic ating that the C terminal tails are important to each protein s function, perhaps in membrane protein complex assembly or protein protein interactions within S. mutans Add itional mutants need to be made t o investigate further the functional domains of the YidC proteins of S. mutans For example, a deletion of the C terminal tail of the YidC1 protein needs to be evaluated to test whether this mutant behaves differently than a complete deletion of yidC1 Additionally, expression of yidC1 fro m the yidC2 promoter and over expression of yidC1 in a yidC2 negative background should be evaluated for stress tolerance.

PAGE 183

183 YidC1 and YidC2 are Involved in ATPase Assembly Experi ments in Chapter 3 showed that YidC1 and YidC2 can both perform in E. coli to insert a functional F 1 F o ATPase, evaluated by ATP hydrolysis assays, PMF assays, and protease accessibility assa ys with the A ssays per formed in S. mutans showed that elimination of either yidC1 or yidC2 resulted in decrease d mem bran e associated ATPase activity In addition, t he activity associate d with the S. mutans yidC1 and yidC2 mutant membranes was not inhibited by the P type i nhibitor orthovanadate, while 13 % of wildtype activity was, indicating that there is less P type activity associated with yidC mutant membranes compared to wildtype This could account for some of the decrease in overall ATPase activity seen in these mutants. T he chimeric YidC1 C2 prote in was able to restore membrane associated activity to a yidC2 mutant. Additionally, AH412 ( yidC2 C ) showed a similar decrease in activity as AH378 ( yidC2 ) indicating the importance of the C terminal tail of YidC2 to ATPase function Future exper iments need to be performed to evaluate membrane ATPas e activity in the presence the F type ATPase inhibitor N, N' dicyclohexyl carbodiimide ( DCCD ) By performing the ATPase assays with mutant membranes in the presence of DCCD, the true proportion of ATPase act ivity attributable to the F 1 F o ATPase can be determined. YidC Mutants Showed Difference s in Membrane Protein Complexes C ompared to Wild T ype NG8 S. mutans There were visible differences in membrane protein complexes between wild type S. mutans and the yid C muta nts (Figures 4 1and 4 5 ). There was an apparent shift in the location and intensity of bands from high molecular we ight complexes present in wild type but absent in both yidC mutants to lower molecular weight complexes in both

PAGE 184

184 yidC mutants that were absent in the wild type This effect was more pronounced in the yidC2 mutant. These differences could indicate a change in the protein expression profile in the yidC mutants and/or they could indicate a defect in the assembly process of multimeric prote in complexes. In the BN PAGE experiments combined with LC MS/MS analysis d ifferences seen in the yidC2 mutant were consistent with an acid to lerance response perhaps stemming from an inability of the mutant to regulate cytoplasmic pH due to defects in the a ssembly of F 1 F o ATPase or oth er affected acid tolerance mechanisms related to the membrane There was also a decrease in ribosomal proteins seen in membranes from both mutants but the effect was more pronounced in the yidC 2 mutant This result is consistent with the C terminal tail of YidC2 having a ribosome binding function that supports co translational translocation. This has been hypothesized as the mechanism behind the dispensability of the SRP pathway in S. mutans Additionally, GAPDH was m ore highly represented in the yidC2 mutant membranes than wild type N G8. W hen extracellular GAPDH activity was measured in whole cells the yidC2 mutant showed increased activity comp ared to wild type NG8 (Figure 4 6) The yidC1 mutant also showed increased extracellular GAPDH activity compared to wildtype, but not to the same extent as the yidC2 mutant. The implications of increased extracellular GAPDH activity remain to be determined. Given the apparent e ffect on membrane as sociated protein complex localization of glycolytic enzymes, f uture exp eriments shoul d be conducted to evaluate further the e ffect s of yidC2 mutation on glycolytic activity including measuring me tabolic end products and by enzymatic assay s Also BN PAGE could be applied to SP20 in which

PAGE 185

185 yidC2 is depleted in a yidC1 negative background, to analyze affects on membrane complexes in the absence of both YidC proteins The Functions of S. mutans YidC1 and YidC2 The purpose of this research was to determine the respective roles of YidC1 and YidC2 in membrane biogenesis in S. mutans A working model is shown in Figure 5 1. Collectively, the knowledge gained from this work a nd the research of others from the Brady lab, indicate a pi votal role of YidC2 in membrane biogenesis and stress tolerance in S mutans that depends on the presence of the C terminal tail Results are consistent with an ability of YidC2 to interact with ribosomes to mediate co tra nslational translocation in coope ration with the SRP pathway (Figure 5 1 A) as evidenced by decreased ribosomes associated with membranes from the yidC2 mutants in BN PAGE /LC MS/MS studies and the ability of YidC2 to complement an Oxa1 mutant in Yeast Additionally, t here is evidence that YidC2 is involved in several pathways that do not or only partially depend on the presence of its C terminal tail, such as the effects seen in competence development which is only partially affected in the yidC2 strain (123) and on the surface adhesin P1 (a sortase substrate) which is not affected at all in the yidC2 strain ( represented by post translational functions in Figure 5 1 B ) There are als o different effects seen between the yidC1 and yidC2 mutants on P1 function, with yidC1 mutant displaying a hyper adherent phenotype, and the yidC2 mutant showi ng a marked decrease is adherence to salivary agglutinin Additionally, b oth y idC mutants affect biofilm for mation, but in different ways. The yidC1 mutant forms biofilms sooner while the biofilms of the yidC2 mutant display a patchy architecture compared to wi ldtype (unpublished personal communication with L. J. Brady ). This

PAGE 186

186 suggests a regulatory role for YidC1 in the secretion of prote ins involved with biofilm formation and perhaps sortase substrates that are covalently linked to the cell wall. Furthermore, t here is also evidence that both YidC1 and YidC2 are involved in the functional assembly of membrane proteins (Figure 5 1 C) reve aled by functional assays with F 1 F o ATPase in E. coli and decreased membrane associated ATPase activity in both yidC mutants in S. mutans Taken together, these resul ts suggest that while YidC2 pla y s a larger role in the biology of S. mutans at least in certain instances YidC1 and YidC2 cooperate in a balanced manner in protein assembly and secretion, thus partially explaining the inability to eliminate both paralogs simultaneously in this species. There are still a number of questions that remain to be answered concerning protein translocation and secretion in S. mutans. There is not a SecB chaperone homolog for post translational targeting in S. mutans This raises the question are proteins post translationally targeted for secretion in S. mutans a n d if so, are there other as yet undiscovered proteins involved? The Streptococci also lack a homolog for SecDF, which in B. subtilis is important for high levels of protein secretion, and in E. coli SecDF are known to associate with YajC. All gram positi ve species encode a y ajC gene; although YajC is not essential in E. coli and deletion has little effect. This raises the question of whether YajC has a more important function in S. mutans and in other g ram positive bacteria than in gram negative species. Also, i t is not known if YajC interact s with the SecY translocon in streptococcal species Addi tionally, the SecYEG translocon has not been characterized in the s treptococci, which makes one speculate that there could be other proteins associated with t his complex in S. mutans that are not found in E. coli Work is underway in the Brady lab to begin to a nswer some of these

PAGE 187

187 questions. This work is vital to understanding the many differences in protein secretion between gram negative bacteria and the medically relevant Streptococcal species.

PAGE 188

188 Figure 5 1. Curren t working m ocation and membrane biogenesis in S. mutans A) YidC2 and the SRP pathway may cooperate in co translational translocation of integral membrane proteins B) The mechanism of post translational targeting in S. mutans is unknown. YidC1 and YidC2 may be in volved in protein translocation, in either a regulatory role or to improve efficiency of secretion or maturation of certain proteins. C) YidC1 and YidC2 are likely involved in the insertion and assembly of multimeric protein complexes. The C terminal t ai l of YidC2 is believed to be required in several of these pathways. A circle around the C terminal tail indicates a variable requirement in the pathway depending on the substrate.

PAGE 189

189 LIST OF REFERENCE S 1. Facklam, R. (2002) What happened to the streptococci: overview of taxonomic and nomenclature changes. Clin Microbiol Rev 15 : 613 30. 2. Doern, C. D. & Burnham, C. A. (2010) It's not easy being green: the viridans group streptococci, wit h a focus on pediatric clinical manifestations. J Clin Microbiol 48 : 3829 35. 3. Nakano, K., Nomura, R., Nakagawa, I., Hamada, S. & Ooshima, T. (2004) Demonstration of Streptococcus mutans with a cell wall polysaccharide specific to a new serotype, k, in th e human oral cavity. J Clin Microbiol 42 : 198 202. 4. Loesche, W. J. (1986) Role of Streptococcus mutans in human dental decay. Microbiol Rev 50 : 353 80. 5. Taubman, M. A. & Nash, D. A. (2006) The scientific and public health imperative for a vaccine against dental caries. Nat Rev Immunol 6 : 555 63. 6. Abranches, J., Zeng, L., Belanger, M., Rodrigues, P. H., Simpson Haidaris, P. J., Akin, D., Dunn, W. A., Jr., Progulske Fox, A. & Burne, R. A. (2009) Invasion of human coronary artery endothelial cells by Strept ococcus mutans OMZ175. Oral Microbiol Immunol 24 : 141 5. 7. Lemos, J. A. & Burne, R. A. (2008) A model of efficiency: stress tolerance by Streptococcus mutans Microbiology 154 : 3247 55. 8. Banas, J. A. & Vickerman, M. M. (2003) Glucan binding proteins of the oral streptococci. Crit Rev Oral Biol Med 14 : 89 99. 9. Jenkinson, H. F. & Demuth, D. R. (1997) Structure, function and immunogenicity of streptococcal antigen I/II polypeptides. Mol Microbiol 23 : 183 90. 10. Ajdic, D., McShan, W. M., McLaughlin, R. E., Savic, G., Chang, J., Carson, M. B., Primeaux, C., Tian, R., Kenton, S., Jia, H., Lin, S., Qian, Y., Li, S., Zhu, H., Najar, F., Lai, H., White, J., Roe, B. A. & Ferretti, J. J. (2002) Genome sequenc e of Streptococcus mutans UA159, a cariogenic dental pathogen. Proc Natl Acad Sci U S A 99 : 14434 9. 11. Margolis, H. C., Zhang, Y. P., Lee, C. Y., Kent, R. L., Jr. & Moreno, E. C. (1999) Kinetics of enamel demineralization in vitro. J Dent Res 78 : 1326 1335 12. Dashper, S. G. & Reynolds, E. C. (1992) pH regulation by Streptococcus mutans J Dent Res 71 : 1159 65.

PAGE 190

190 13. Bender, G. R., Thibodeau, E. A. & Marquis, R. E. (1985) Reduction of acidurance of streptococcal growth and glycolysis by fluoride and gramicidi n. J Dent Res 64 : 90 95. 14. Len, A. C., Harty, D. W. & Jacques, N. A. (2004) Proteome analysis of Streptococcus mutans metabolic phenotype during acid tolerance. Microbiology 150 : 1353 66. 15. Len, A. C., Harty, D. W. & Jacques, N. A. (2004) Stress responsi ve proteins are upregulated in Streptococcus mutans during acid tolerance. Microbiology 150 : 1339 51. 16. Hamilton, I. R. & Buckley, N. D. (1991) Adaptation by Streptococcus mutans to acid tolerance. Oral Microbiol Immunol 6 : 65 71. 17. Fozo, E. M. & Quivey, R. G., Jr. (2004) The fabM gene product of Streptococcus mutans is responsible for the synthesis of monounsaturated fatty acids and is necessary for survival at low pH. J Bacteriol 186 : 4152 8. 18. Fozo, E. M. & Quivey, R. G., Jr. (2004) Shifts in the memb rane fatty acid profile of Streptococcus mutans enhance survival in acidic environments. Appl Environ Microbiol 70 : 929 36. 19. Quivey, R. G., Kuhnert, W. L. & Hahn, K. (2001) Genetics of acid adaptation in oral streptococci. Crit Rev Oral Biol Med 12 : 301 1 4. 20. Carlsson J. (1986) Textbook of Cariology. Munksgaard, Copenhagen 21. Senior, A. E., Nadanaciva, S. & Weber, J. (2002) The molecular mechanism of ATP synthesis by F1F0 ATP synthase. Biochim Biophys Acta 1553 : 188 211. 22. Sheng, J. & Marquis, R. E. (2006) Enhanced acid resistance of oral streptococci at lethal pH values associated with acid tolerant catabolism and with ATP synthase activity. FEMS Microbiol Lett 262 : 93 8. 23. Bender, G. R., Sutton, S. V. & Marquis, R. E. (1986) Acid tolerance, proton permeabilities, and membrane ATPases of oral streptococci. Infect Immun 53 : 331 8. 24. Sutton, S. V. & Marquis, R. E. (1987) Membrane associated and solubilized ATPases of Streptococcus mutans and Streptococcus sanguis J Dent Res 66 : 1095 8.

PAGE 191

191 25. Kuhnert, W L., Zheng, G., Faustoferri, R. C. & Quivey, R. G., Jr. (2004) The F ATPase operon promoter of Streptococcus mutans is transcriptionally regulated in response to external pH. J Bacteriol 186 : 8524 8. 26. Magalhaes, P. P., Paulino, T. P., Thedei, G., Jr. & Ciancaglini, P. (2005) Kinetic characterization of P type membrane ATPase from Streptococcus mutans Comp Biochem Physiol B Biochem Mol Biol 140 : 589 97. 27. Gong, Y., Tian, X. L., Sutherland, T., Sisson, G., Mai, J., Ling, J. & Li, Y. H. (2009) Global transcriptional analysis of acid inducible genes in Streptococcus mutans : multiple two component systems involved in acid adaptation. Microbiology 155 : 3322 32. 28. Yi, L., Celebi, N., Chen, M. & Dalbey, R. E. (2004) Sec/SRP require ments and energetics of membrane insertion of subunits a, b, and c of the Escherichia coli F1F0 ATP synthase. J Biol Chem 279 : 39260 7. 29. Kol, S., Majczak, W., Heerlien, R., van der Berg, J. P., Nouwen, N. & Driessen, A. J. (2009) Subunit a of the F(1)F(0 ) ATP synthase requires YidC and SecYEG for membrane insertion. J Mol Biol 390 : 893 901. 30. van der Laan, M., Bechtluft, P., Kol, S., Nouwen, N. & Driessen, A. J. (2004) F1F0 ATP synthase subunit c is a substrate of the novel YidC pathway for membrane prot ein biogenesis. J Cell Biol 165 : 213 22. 31. Hasona, A., Crowley, P. J., Levesque, C. M., Mair, R. W., Cvitkovitch, D. G., Bleiweis, A. S. & Brady, L. J. (2005) Streptococcal viability and diminished stress tolerance in mutants lacking the signal recognitio n particle pathway or YidC2. Proc Natl Acad Sci U S A 102 : 17466 71. 32. Ajdic, D. & Pham, V. T. (2007) Global transcriptional analysis of Streptococcus mutans sugar transporters using microarrays. J Bacteriol 189 : 5049 59. 33. Postma, P. W., Lengeler, J. W. & Jacobson, G. R. (1993) Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiol Rev 57 : 543 94. 34. Zeng, L. & Burne, R. A. (2010) Seryl phosphorylated HPr Regulates CcpA Independent Carbon Catabolite Repression in Conjunction w ith PTS Permeases in Streptococcus mutans Mol Microbiol 35. Titgemeyer, F. & Hillen, W. (2002) Global control of sugar metabolism: a gram positive solution. Antonie Van Leeuwenhoek 82 : 59 71.

PAGE 192

192 36. Korithoski, B., Levesque, C. M. & Cvitkovitch, D. G. (2008) The involvement of the pyruvate dehydrogenase E1alpha subunit, in Streptococcus mutans acid tolerance. FEMS Microbiol Lett 289 : 13 9. 37. Carlsson, J., Kujala, U. & Edlund, M. B. (1985) Pyruvate dehydrogenase activity in Streptococcus mutans Infect Immun 49 : 674 8. 38. Mascher, T., Helmann, J. D. & Unden, G. (2006) Stimulus perception in bacterial signal transducing histidine kinases. Microbiol Mol Biol Rev 70 : 910 38. 39. Biswas, I., Drake, L., Erkina, D. & Biswas, S. (2008) Involvement of sensor kinases i n the stress tolerance response of Streptococcus mutans J Bacteriol 190 : 68 77. 40. Senadheera, D., Krastel, K., Mair, R., Persadmehr, A., Abranches, J., Burne, R. A. & Cvitkovitch, D. G. (2009) Inactivation of VicK affects acid production and acid surviva l of Streptococcus mutans J Bacteriol 191 : 6415 24. 41. bChong, P., Drake, L. & Biswas, I. (2008) LiaS regulates virulence factor expression in Streptococcus mutans Infect Immun 76 : 3093 9. 42. aChong, P., Drake, L. & Biswas, I. (2008) Modulation of covR e xpression in Streptococcus mutans UA159. J Bacteriol 190 : 4478 88. 43. Banu, L. D., Conrads, G., Rehrauer, H., Hussain, H., Allan, E. & van der Ploeg, J. R. (2010) The Streptococcus mutans serine/threonine kinase, PknB, regulates competence development, bac teriocin production, and cell wall metabolism. Infect Immun 78 : 2209 20. 44. Mascher, T. (2006) Intramembrane sensing histidine kinases: a new family of cell envelope stress sensors in Firmicutes bacteria. FEMS Microbiol Lett 264 : 133 44. 45. Suntharalingam, P., Senadheera, M. D., Mair, R. W., Levesque, C. M. & Cvitkovitch, D. G. (2009) The LiaFSR system regulates the cell envelope stress response in Streptococcus mutans J Bacteriol 191 : 2973 84. 46. Yuan, J., Zweers, J. C., van Dijl, J. M. & D albey, R. E. (2010) Protein transport across and into cell membranes in bacteria and archaea. Cell Mol Life Sci 67 : 179 99. 47. Driessen, A. J. & Nouwen, N. (2008) Protein translocation across the bacterial cytoplasmic membrane. Annu Rev Biochem 77 : 643 67.

PAGE 193

193 48. Papanikou, E., Karamanou, S. & Economou, A. (2007) Bacterial protein secretion through the translocase nanomachine. Nat Rev Microbiol 5 : 839 51. 49. Hoffmann, A., Bukau, B. & Kramer, G. (2010) Structure and function of the molecular chaperone Trigger Fa ctor. Biochim Biophys Acta 1803 : 650 61. 50. Beck, K., Wu, L. F., Brunner, J. & Muller, M. (2000) Discrimination between SRP and SecA/SecB dependent substrates involves selective recognition of nascent chains by SRP and trigger factor. EMBO J 19 : 134 43. 51 Duong, F. & Wickner, W. (1997) Distinct catalytic roles of the SecYE, SecG and SecDFyajC subunits of preprotein translocase holoenzyme. Embo J 16 : 2756 68. 52. du Plessis, D. J., Nouwen, N. & Driessen, A. J. (2010) The Sec translocase. Biochim Biophys Act a 53. Pogliano, J. A. & Beckwith, J. (1994) SecD and SecF facilitate protein export in Escherichia coli EMBO J 13 : 554 61. 54. Bolhuis, A., Broekhuizen, C. P., Sorokin, A., van Roosmalen, M. L., Venema, G., Bron, S., Quax, W. J. & van Dijl, J. M. (1998) S ecDF of Bacillus subtilis a molecular Siamese twin required for the efficient secretion of proteins. J Biol Chem 273 : 21217 24. 55. Yamane, K., Bunai, K. & Kakeshita, H. (2004) Protein traffic for secretion and related machinery of Bacillus subtilis Biosc i Biotechnol Biochem 68 : 2007 23. 56. Murphy, C. K. & Beckwith, J. (1994) Residues essential for the function of SecE, a membrane component of the Escherichia coli secretion apparatus, are located in a conserved cytoplasmic region. Proc Natl Acad Sci U S A 91 : 2557 61. 57. Nakamura, K., Yahagi, S., Yamazaki, T. & Yamane, K. (1999) Bacillus subtilis histone like protein, HBsu, is an integral component of a SRP like particle that can bind the Alu domain of small cytoplasmic RNA. J Biol Chem 274 : 13569 76. 58. Ho nda, K., Nakamura, K., Nishiguchi, M. & Yamane, K. (1993) Cloning and characterization of a Bacillus subtilis gene encoding a homolog of the 54 kilodalton subunit of mammalian signal recognition particle and Escherichia coli Ffh. J Bacteriol 175 : 4885 94.

PAGE 194

194 5 9. Swaving, J., van Wely, K. H. & Driessen, A. J. (1999) Preprotein translocation by a hybrid translocase composed of Escherichia coli and Bacillus subtilis subunits. J Bacteriol 181 : 7021 7. 60. Gutierrez, J. A., Crowley, P. J., Brown, D. P., Hillman, J. D., Youngman, P. & Bleiweis, A. S. (1996) Insertional mutagenesis and recovery of interrupted genes of Streptococcus mutans by using transposon Tn917: preliminary characterization of mutants displ aying acid sensitivity and nutritional requirements. J Bacteriol 178 : 4166 75. 61. Gutierrez, J. A., Crowley, P. J., Cvitkovitch, D. G., Brady, L. J., Hamilton, I. R., Hillman, J. D. & Bleiweis, A. S. (1999) Streptococcus mutans ffh a gene encoding a homol ogue of the 54 kDa subunit of the signal recognition particle, is involved in resistance to acid stress. Microbiology 145 ( Pt 2) : 357 66. 62. Rosch, J. W., Vega, L. A., Beyer, J. M., Lin, A. & Caparon, M. G. (2008) The signal recognition particle pathway i s required for virulence in Streptococcus pyogenes Infect Immun 76 : 2612 9. 63. Rigel, N. W. & Braunstein, M. (2008) A new twist on an old pathway -accessory Sec [corrected] systems. Mol Microbiol 69 : 291 302. 64. Bensing, B. A. & Sullam, P. M. (2002) An ac cessory sec locus of Streptococcus gordonii is required for export of the surface protein GspB and for normal levels of binding to human platelets. Mol Microbiol 44 : 1081 94. 65. Bensing, B. A., Siboo, I. R. & Sullam, P. M. (2007) Glycine residues in the hy drophobic core of the GspB signal sequence route export toward the accessory Sec pathway. J Bacteriol 189 : 3846 54. 66. Takamatsu, D., Bensing, B. A. & Sullam, P. M. (2004) Genes in the accessory sec locus of Streptococcus gordonii have three functionally distinct effects on the expression of the platelet binding protein GspB. Mol Microbiol 52 : 189 203. 67. Rosch, J. W. & Caparon, M. G. (2005) The ExPortal: an organelle dedicated to the biogenesis of secreted proteins in Streptococcus pyogenes Mol Microbiol 58 : 959 68. 68. Rosch, J. & Caparon, M. (2004) A microdomain for protein secretion in Gram positive bacteria. Science 304 : 1513 5. 69. Hu, P., Bian, Z., Fan, M., Huang, M. & Zhang, P. (2008) Sec translocase and sortase A are colocali sed in a locus in the cytoplasmic membrane of Streptococcus mutans Arch Oral Biol 53 : 150 4.

PAGE 195

195 70. Carlsson, F., Stalhammar Carlemalm, M., Flardh, K., Sandin, C., Carlemalm, E. & Lindahl, G. (2006) Signal sequence directs localized secretion of bacterial sur face proteins. Nature 442 : 943 6. 71. Campo, N., Tjalsma, H., Buist, G., Stepniak, D., Meijer, M., Veenhuis, M., Westermann, M., Muller, J. P., Bron, S., Kok, J., Kuipers, O. P. & Jongbloed, J. D. (2004) Subcellular sites for bacterial protein export. Mol M icrobiol 53 : 1583 99. 72. Phillips, G. J. & Silhavy, T. J. (1992) The E. coli ffh gene is necessary for viability and efficient protein export. Nature 359 : 744 6. 73. Kremer, B. H., van der Kraan, M., Crowley, P. J., Hamilton, I. R., Brady, L. J. & Bleiweis, A. S. (2001) Characterization of the sat operon in Streptococcus mutans : evidence for a role of Ffh in acid tolerance. J Bacteriol 183 : 2543 52. 74. Crowley, P. J., Svensater, G., Snoep, J. L., Bleiweis, A. S. & Brady, L. J. (2004) An ffh mutant of Strepto coccus mutans is viable and able to physiologically adapt to low pH in continuous culture. FEMS Microbiol Lett 234 : 315 24. 75. Hasona, A., Zuobi Hasona, K., Crowley, P. J., Abranches, J., Ruelf, M. A., Bleiweis, A. S. & Brady, L. J. (2007) Membrane composition changes and physiological adaptation by Streptococcus mutans signal recognition particle pathway mutants. J Bacteriol 189 : 1219 30. 76. Li, Y. H., Tang, N., Aspiras, M. B., Lau, P. C., Lee, J. H., Ellen, R. P. & Cvi tkovitch, D. G. (2002) A quorum sensing signaling system essential for genetic competence in Streptococcus mutans is involved in biofilm formation. J Bacteriol 184 : 2699 708. 77. Zhang, Y. J., Tian, H. F. & Wen, J. F. (2009) The evolution of YidC/Oxa/Alb3 f amily in the three domains of life: a phylogenomic analysis. BMC Evol Biol 9 : 137. 78. Bonnefoy, N., Fiumera, H. L., Dujardin, G. & Fox, T. D. (2009) Roles of Oxa1 related inner membrane translocases in assembly of respiratory chain complexes. Biochim Bioph ys Acta 1793 : 60 70. 79. van der Laan, M., Urbanus, M. L., Ten Hagen Jongman, C. M., Nouwen, N., Oudega, B., Harms, N., Driessen, A. J. & Luirink, J. (2003) A conserved function of YidC in the biogenesis of respiratory chain complexes. Proc Natl Acad Sci U S A 100 : 5801 6. 80. Kiefer, D. & Kuhn, A. (2007) YidC as an essential and multifunctional component in membrane protein assembly. Int Rev Cytol 259 : 113 38.

PAGE 196

196 81. Yen, M. R., Harley, K. T., Tseng, Y. H. & Saier, M. H., Jr. (2001) Phylogenetic and structural a nalyses of the oxa1 family of protein translocases. FEMS Microbiol Lett 204 : 223 31. 82. Bonnefoy, N., Chalvet, F., Hamel, P., Slonimski, P. P. & Dujardin, G. (1994) OXA1, a Saccharomyces cerevisiae nuclear gene whose sequence is conserved from prokaryotes to eukaryotes controls cytochrome oxidase biogenesis. J Mol Biol 239 : 201 12. 83. Yi, L. & Dalbey, R. E. (2005) Oxa1/Alb3/YidC system for insertion of membrane proteins in mitochondria, chloroplasts and bacteria (review). Mol Membr Biol 22 : 101 11. 84. Stuar t, R. (2002) Insertion of proteins into the inner membrane of mitochondria: the role of the Oxa1 complex. Biochim Biophys Acta 1592 : 79 87. 85. Jia, L., Dienhart, M., Schramp, M., McCauley, M., Hell, K. & Stuart, R. A. (2003) Yeast Oxa1 interacts with mitoc hondrial ribosomes: the importance of the C terminal region of Oxa1. Embo J 22 : 6438 47. 86. Saracco, S. A. & Fox, T. D. (2002) Cox18p is required for export of the mitochondrially encoded Saccharomyces cerevisiae Cox2p C tail and interacts with Pnt1p and M ss2p in the inner membrane. Mol Biol Cell 13 : 1122 31. 87. Funes, S., Nargang, F. E., Neupert, W. & Herrmann, J. M. (2004) The Oxa2 Protein of Neurospora crassa Plays a Critical Role in the Biogenesis of Cytochrome Oxidase and Defines a Ubiquitous Subbranch of the Oxa1/YidC/Alb3 Protein Family. Mol. Biol. Cell 15 : 1853 1861. 88. Gaisne, M. & Bonnefoy, N. (2006) The COX18 gene, involved in mitochondrial biogenesis, is functionally conserved and tightly regulated in humans and fission yeast. FEMS Yeast Res 6 : 86 9 82. 89. Samuelson, J. C., Chen, M., Jiang, F., Moller, I., Wiedmann, M., Kuhn, A., Phillips, G. J. & Dalbey, R. E. (2000) YidC mediates membrane protein insertion in bacteria. Nature 406 : 637 41. 90. Saaf, A., Monne, M., de Gier, J. W. & von Heijne, G. (1 998) Membrane topology of the 60 kDa Oxa1p homologue from Escherichia coli J Biol Chem 273 : 30415 8. 91. Jiang, F., Chen, M., Yi, L., de Gier, J. W., Kuhn, A. & Dalbey, R. E. (2003) Defining the regions of Escherichia coli YidC that contribute to activity. J Biol Chem 278 : 48965 72.

PAGE 197

197 92. van der Laan, M., Houben, E. N., Nouwen, N., Luirink, J. & Driessen, A. J. (2001) Reconstitution of Sec dependent membrane protein insertion: nascent FtsQ interacts with YidC in a SecYEG dependent manner. EMBO Rep 2 : 519 23. 93. Urbanus, M. L., Froderberg, L., Drew, D., Bjork, P., de Gier, J. W., Brunner, J., Oudega, B. & Luirink, J. (2002) Targeting, insertion, and localization of Escherichia coli YidC. J Biol Chem 277 : 12718 23. 94. Celebi, N., Yi, L., Facey, S. J., Kuhn, A. & Dalbey, R. E. (2006) Membrane biogenesis of subunit II of cytochrome bo oxidase: contrasting requirements for insertion of N terminal and C terminal domains. J Mol Biol 357 : 1428 36. 95. Chen, M., Xie, K., Nouwen, N., Driessen, A. J. & Dalbey, R. E. (2003) Conditional lethal mutations separate the M13 procoat and Pf3 coat functions of YidC: different YIDC structural requirements for membrane protein insertion. J Biol Chem 278 : 23295 300. 96. Kol, S., Nouwen, N. & Driessen, A. J. (2008) T he charge distribution in the cytoplasmic loop of subunit C of the F1F0 ATPase is a determinant for YidC targeting. J Biol Chem 283 : 9871 7. 97. Facey, S. J., Neugebauer, S. A., Krauss, S. & Kuhn, A. (2007) The mechanosensitive channel protein MscL is targe ted by the SRP to the novel YidC membrane insertion pathway of Escherichia coli J Mol Biol 365 : 995 1004. 98. Price, C. E., Kocer, A., Kol, S., van der Berg, J. P. & Driessen, A. J. (2011) In vitro synthesis and oligomerization of the mechanosensitive chan nel of large conductance, MscL, into a functional ion channel. FEBS Lett 585 : 249 54. 99. Wagner, S., Pop, O. I., Haan, G. J., Baars, L., Koningstein, G., Klepsch, M. M., Genevaux, P., Luirink, J. & de Gier, J. W. (2008) Biogenesis of MalF and the MalFGK(2) maltose transport complex in Escherichia coli requires YidC. J Biol Chem 283 : 17881 90. 100. Nagamori, S., Smirnova, I. N. & Kaback, H. R. (2004) Role of YidC in folding of polytopic membrane proteins. J Cell Biol 165 : 53 62. 101. Beck, K., Eisner, G., Tres cher, D., Dalbey, R. E., Brunner, J. & Muller, M. (2001) YidC, an assembly site for polytopic Escherichia coli membrane proteins located in immediate proximity to the SecYE translocon and lipids. EMBO Rep 2 : 709 14.

PAGE 198

198 102. Sundberg, E., Slagter, J. G., Fridbo rg, I., Cleary, S. P., Robinson, C. & Coupland, G. (1997) ALBINO3, an Arabidopsis nuclear gene essential for chloroplast differentiation, encodes a chloroplast protein that shows homology to proteins present in bacterial membranes and yeast mitochondria. P lant Cell 9 : 717 30. 103. Moore, M., Harrison, M. S., Peterson, E. C. & Henry, R. (2000) Chloroplast Oxa1p homolog albino3 is required for post translational integration of the light harvesting chlorophyll binding protein into thylakoid membranes. J Biol Ch em 275 : 1529 32. 104. Gerdes, L., Bals, T., Klostermann, E., Karl, M., Philippar, K., Hunken, M., Soll, J. & Schunemann, D. (2006) A second thylakoid membrane localized Alb3/OxaI/YidC homologue is involved in proper chloroplast biogenesis in Arabidopsis tha liana J Biol Chem 281 : 16632 42. 105. Gohre, V., Ossenbuhl, F., Crevecoeur, M., Eichacker, L. A. & Rochaix, J. D. (2006) One of two alb3 proteins is essential for the assembly of the photosystems and for cell survival in Chlamydomonas Plant Cell 18 : 1454 6 6. 106. Aldridge, C., Cain, P. & Robinson, C. (2009) Protein transport in organelles: Protein transport into and across the thylakoid membrane. FEBS J 276 : 1177 86. 107. Schunemann, D. (2007) Mechanisms of protein import into thylakoids of chloroplasts. Bio l Chem 388 : 907 15. 108. Pool, M. R. (2005) Signal recognition particles in chloroplasts, bacteria, yeast and mammals (review). Mol Membr Biol 22 : 3 15. 109. Lewis, N. E., Marty, N. J., Kathir, K. M., Rajalingam, D., Kight, A. D., Daily, A., Kumar, T. K., He nry, R. L. & Goforth, R. L. (2010) A dynamic cpSRP43 Albino3 interaction mediates translocase regulation of chloroplast signal recognition particle (cpSRP) targeting components. J Biol Chem 285 : 34220 30. 110. Falk, S., Ravaud, S., Koch, J. & Sinning, I. (2 010) The C terminus of the Alb3 membrane insertase recruits cpSRP43 to the thylakoid membrane. J Biol Chem 285 : 5954 62. 111. Klostermann, E., Droste Gen Helling, I., Carde, J. P. & Schunemann, D. (2002) The thylakoid membrane protein ALB3 associates with t he cpSecY translocase in Arabidopsis thaliana Biochem J 368 : 777 81.

PAGE 199

199 112. Benz, M., Bals, T., Gugel, I. L., Piotrowski, M., Kuhn, A., Schunemann, D., Soll, J. & Ankele, E. (2009) Alb4 of Arabidopsis Promotes Assembly and Stabilization of a Non Chlorophyll Binding Photosynthetic Complex, the CF1CF0 ATP Synthase. Mol Plant 2 : 1410 24. 113. Sutcliffe, I. C. & Russell, R. R. (1995) Lipoproteins of gram positive bacter ia. J Bacteriol 177 : 1123 8. 114. Murakami, T., Haga, K., Takeuchi, M. & Sato, T. (2002) Analysis of the Bacillus subtilis spoIIIJ gene and its Paralogue gene, yqjG. J Bacteriol 184 : 1998 2004. 115. Tjalsma, H., Bron, S. & van Dijl, J. M. (2003) Complementar y Impact of Paralogous Oxa1 like Proteins of Bacillus subtilis on Post translocational Stages in Protein Secretion. J. Biol. Chem. 278 : 15622 15632. 116. Saller, M. J., Fusetti, F. & Driessen, A. J. (2009) Bacillus subtilis SpoIIIJ and YqjG function in memb rane protein biogenesis. J Bacteriol 191 : 6749 57. 117. Saller, M. J., Otto, A., Berrelkamp Lahpor, G. A., Becher, D., Hecker, M. & Driessen, A. J. (2011) Bacillus subtilis YqjG is required for genetic competence development. Proteomics 11 : 270 82. 118. Rubi o, A., Jiang, X. & Pogliano, K. (2005) Localization of translocation complex components in Bacillus subtilis : enrichment of the signal recognition particle receptor at early sporulation septa. J Bacteriol 187 : 5000 2. 119. Chiba, S., Lamsa, A. & Pogliano, K (2009) A ribosome nascent chain sensor of membrane protein biogenesis in Bacillus subtilis EMBO J 28 : 3461 75. 120. Sarker, S., Rudd, K. E. & Oliver, D. (2000) Revised translation start site for secM defines an atypical signal peptide that regulates Esch erichia coli secA expression. J Bacteriol 182 : 5592 5. 121. Brady, L. J., Piacentini, D. A., Crowley, P. J. & Bleiweis, A. S. (1991) Identification of monoclonal antibody binding domains within antigen P1 of Streptococcus mutans and cross reactivity with re lated surface antigens of oral streptococci. Infect Immun 59 : 4425 35. 122. Oli, M. W., McArthur, W. P. & Brady, L. J. (2006) A whole cell BIAcore assay to evaluate P1 mediated adherence of Streptococcus mutans to human salivary agglutinin and inhibition by specific antibodies. J Microbiol Methods 65 : 503 11.

PAGE 200

200 123. Funes, S., Hasona, A., Bauerschmitt, H., Grubbauer, C., Kauff, F., Collins, R., Crowley, P. J., Palmer, S. R., Brady, L. J. & Herrmann, J. M. (2009) Independent gene duplications of the YidC/Oxa/Alb 3 family enabled a specialized cotranslational function. Proc Natl Acad Sci U S A 106 : 6656 61. 124. Jones, S. E., Lloyd, L. J., Tan, K. K. & Buck, M. (2003) Secretion defects that activate the phage shock response of Escherichia coli J Bacteriol 185 : 6707 11. 125. Antiporta, M. H. & Dunny, G. M. (2002) ccfA, the genetic determinant for the cCF10 peptide pheromone in Enterococcus faecalis OG1RF. J Bacteriol 184 : 1155 62. 126. Terpe, K. (2006) Overview of bacterial expression systems for heterologous protein production: from molecular and biochemical fundamentals to commercial systems. Appl Microbiol Biotechnol 72 : 211 22. 127. Baars, L., Wagner, S., Wickstrom, D., Klepsch, M., Y tterberg, A. J., van Wijk, K. J. & de Gier, J. W. (2008) Effects of SecE depletion on the inner and outer membrane proteomes of Escherichia coli J Bacteriol 190 : 3505 25. 128. de Gier, J. W., Mansournia, P., Valent, Q. A., Phillips, G. J., Luirink, J. & vo n Heijne, G. (1996) Assembly of a cytoplasmic membrane protein in Escherichia coli is dependent on the signal recognition particle. FEBS Lett 399 : 307 9. 129. Herskovits, A. A., Seluanov, A., Rajsbaum, R., ten Hagen Jongman, C. M., Henrichs, T., Bochkareva, E. S., Phillips, G. J., Probst, F. J., Nakae, T., Ehrmann, M., Luirink, J. & Bibi, E. (2001) Evidence for coupling of membrane targeting and function of the signal recognition particle (SRP) receptor FtsY. EMBO Rep 2 : 1040 6. 130. Henner, D. J. (1990) Indu cible expression of regulatory genes in Bacillus subtilis Methods Enzymol 185 : 223 8. 131. Eichenbaum, Z., Federle, M. J., Marra, D., de Vos, W. M., Kuipers, O. P., Kleerebezem, M. & Scott, J. R. (1998) Use of the lactococcal nisA promoter to regulate gene expression in gram positive bacteria: comparison of induction level and promoter strength. Appl Environ Microbiol 64 : 2763 9. 132. Bryan, E. M., Bae, T., Kleerebezem, M. & Dunny, G. M. (2000) Improved vectors for nisin controlled expression in gram positiv e bacteria. Plasmid 44 : 183 90.

PAGE 201

201 133. Zeng, L. & Burne, R. A. (2009) Transcriptional regulation of the cellobiose operon of Streptococcus mutans J Bacteriol 191 : 2153 62. 134. Van den Berg, B., Clemons, W. M., Jr., Collinson, I., Modis, Y., Hartmann, E., Har rison, S. C. & Rapoport, T. A. (2004) X ray structure of a protein conducting channel. Nature 427 : 36 44. 135. Heckman, K. L. & Pease, L. R. (2007) Gene splicing and mutagenesis by PCR driven overlap extension. Nat Protoc 2 : 924 32. 136. Wen, Z. T. & Burne, R. A. (2001) Construction of a new integration vector for use in Streptococcus mutans Plasmid 45 : 31 6. 137. LeBlanc, D. J., Lee, L. N. & Abu Al Jaibat, A. (1992) Molecular, genetic, and functional analysis of the basic replicon of pVA380 1, a plasmid of o ral streptococcal origin. Plasmid 28 : 130 45. 138. Terleckyj, B., Willett, N. P. & Shockman, G. D. (1975) Growth of several cariogenic strains of oral streptococci in a chemically defined medium. Infect Immun 11 : 649 55. 139. Price, C. E., Otto, A., Fusetti, F., Becher, D., Hecker, M. & Driessen, A. J. (2010) Differential effect of YidC depletion on the membrane proteome of Escherichia coli under aerobic and anaerobic growth conditions. Proteomics 10 : 3235 47. 140. Abranches, J., Chen, Y. Y. & Burne, R. A. (20 03) Characterization of Streptococcus mutans strains deficient in EIIAB Man of the sugar phosphotransferase system. Appl Environ Microbiol 69 : 4760 9. 141. Abranches, J., Candella, M. M., Wen, Z. T., Baker, H. V. & Burne, R. A. (2006) Different roles of EII ABMan and EIIGlc in regulation of energy metabolism, biofilm development, and competence in Streptococcus mutans J Bacteriol 188 : 3748 56. 142. Dong, Y., Palmer, S. R., Hasona, A., Nagamori, S., Kaback, H. R., Dalbey, R. E. & Brady, L. J. (2008) Functional overlap but lack of complete cross complementation of Streptococcus mutans and Escherichia coli YidC orthologs. J Bacteriol 190 : 2458 69. 143. Buckley, N. D., Lee, L. N. & LeBlanc, D. J. (1995) Use of a novel mobilizable vector to inactivate the scrA gene of Streptococcus sobrinus by allelic replacement. J Bacteriol 177 : 5028 34. 144. van Bloois, E., Jan Haan, G., de Gier, J. W., Oudega, B. & Luirink, J. (2004) F(1)F(0) ATP synthase subunit c is targeted by the SRP to YidC in the E. coli inner membrane. FEBS Lett 576 : 97 100.

PAGE 202

202 145. Kol, S., Turrell, B. R., de Keyzer, J., van der Laan, M., Nouwen, N. & Driessen, A. J. (2006) YidC mediated membrane insertion of assembly mutants o f subunit c of the F1F0 ATPase. J Biol Chem 281 : 29762 8. 146. Ricci, M. L., Manganelli, R., Berneri, C., Orefici, G. & Pozzi, G. (1994) Electrotransformation of Streptococcus agalactiae with plasmid DNA. FEMS Microbiol Lett 119 : 47 52. 147. Samuelson, J. C. Jiang, F., Yi, L., Chen, M., de Gier, J. W., Kuhn, A. & Dalbey, R. E. (2001) Function of YidC for the insertion of M13 procoat protein in Escherichia coli : translocation of mutants that show differences in their membrane potential dependence and Sec requ irement. J Biol Chem 276 : 34847 52. 148. Bhatt, D., Cole, S. P., Grabar, T. B., Claggett, S. B. & Cain, B. D. (2005) Manipulating the length of the b subunit F1 binding domain in F1F0 ATP synthase from Escherichia coli J Bioenerg Biomembr 37 : 67 74. 149. Fi ske, C. H. & Subbarow, Y. (1925) THE COLORIMETRIC DETERMINATION OF PHOSPHORUS. Journal of Biological Chemistry 66 : 375 400. 150. Sorgen, P. L., Bubb, M. R., McCormick, K. A., Edison, A. S. & Cain, B. D. (1998) Formation of the b subunit dimer is necessary f or interaction with F1 ATPase. Biochemistry 37 : 923 32. 151. Stack, A. E. & Cain, B. D. (1994) Mutations in the delta subunit influence the assembly of F1F0 ATP synthase in Escherichia coli J Bacteriol 176 : 540 2. 152. Welch, A. K., Claggett, S. B. & Cain, B. D. (2008) The b (arg36) contributes to efficient coupling in F(1)F (O) ATP synthase in Escherichia coli J Bioenerg Biomembr 40 : 1 8. 153. du Plessis, D. J., Nouwen, N. & Driessen, A. J. (2006) Subunit a of cytochrome o oxidase requires both YidC and SecYEG for membrane insertion. J Biol Chem 281 : 12248 52. 154. Nieuwenhuis, F. J., Kanner, B. I., Gutnick, D. L., Postma, P. W. & van Dam, K. (1973) Energy conservation in memb ranes of mutants of Escherichia coli defective in oxidative phosphorylation. Biochim Biophys Acta 325 : 62 71. 155. Schagger, H. & von Jagow, G. (1991) Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Anal Biochem 199 : 223 31.

PAGE 203

203 156. Neff, D. & Dencher, N. A. (1999) Purification of multisubunit membrane protein complexes: isolation of chloroplast FoF1 ATP synthase, CFo and CF1 by blue native electrophoresis. Biochem Biophys Res Commun 259 : 569 75. 157. Cline, K & Mori, H. (2001) Thylakoid DeltapH dependent precursor proteins bind to a cpTatC Hcf106 complex before Tha4 dependent transport. J Cell Biol 154 : 719 29. 158. Krause, F. (2006) Detection and analysis of protein protein interactions in organellar and prok aryotic proteomes by native gel electrophoresis: (Membrane) protein complexes and supercomplexes. Electrophoresis 27 : 2759 81. 159. Braun, R. J., Kinkl, N., Beer, M. & Ueffing, M. (2007) Two dimensional electrophoresis of membrane proteins. Anal Bioanal Che m 389 : 1033 45. 160. Dresler, J., Klimentova, J. & Stulik, J. (2011) Bacterial protein complexes investigation using blue native PAGE. Microbiol Res 166 : 47 62. 161. Stenberg, F., Chovanec, P., Maslen, S. L., Robinson, C. V., Ilag, L. L., von Heijne, G. & Daley, D. O. (2005) Protein complexes of the Escherichia coli cell envelope. J Biol Chem 280 : 34409 19. 162. Aivaliotis, M., Karas, M. & Tsiotis, G. (2007) An alternative strategy for the membrane proteome analysis of the green sulfur bacterium Chlorobium tepidum using blue native PAGE and 2 D PAGE on purified membranes. J Proteome Res 6 : 1048 58. 163. Tsirogianni, E., Aivaliotis, M., Papasotiriou, D. G., Karas, M. & Tsiotis, G. (2006) Identification of inducible protein complexes in the phenol degrader Pseu domonas sp. strain phDV1 by blue native gel electrophoresis and mass spectrometry. Amino Acids 30 : 63 72. 164. Heuberger, E. H., Veenhoff, L. M., Duurkens, R. H., Friesen, R. H. & Poolman, B. (2002) Oligomeric state of membrane transport proteins analyzed w ith blue native electrophoresis and analytical ultracentrifugation. J Mol Biol 317 : 591 600. 165. Chassy, B. M. (1976) A gentle method for the lysis of oral streptococci. Biochem Biophys Res Commun 68 : 603 8. 166. Wittig, I., Braun, H. P. & Schagger, H. (200 6) Blue native PAGE. Nat Protoc 1 : 418 28. 167. Keller, A., Nesvizhskii, A. I., Kolker, E. & Aebersold, R. (2002) Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Anal Chem 74 : 5383 92.

PAGE 204

204 168. N esvizhskii, A. I., Keller, A., Kolker, E. & Aebersold, R. (2003) A statistical model for identifying proteins by tandem mass spectrometry. Anal Chem 75 : 4646 58. 169. Schagger, H. (2006) Tricine SDS PAGE. Nat Protoc 1 : 16 22. 170. Akiyama, Y. & Ito, K. (1985 ) The SecY membrane component of the bacterial protein export machinery: analysis by new electrophoretic methods for integral membrane proteins. EMBO J 4 : 3351 6. 171. Bessonneau, P., Besson, V., Collinson, I. & Duong, F. (2002) The SecYEG preprotein transl ocation channel is a conformationally dynamic and dimeric structure. EMBO J 21 : 995 1003. 172. Rath, A., Nadeau, V. G., Poulsen, B. E., Ng, D. P. & Deber, C. M. (2010) Novel hydrophobic standards for membrane protein molecular weight determinations via sodi um dodecyl sulfate polyacrylamide gel electrophoresis. Biochemistry 49 : 10589 91. 173. Tziatzios, C., Schubert, D., Lotz, M., Gundogan, D., Betz, H., Schagger, H., Haase, W., Duong, F. & Collinson, I. (2004) The bacterial protein translocation complex: SecY EG dimers associate with one or two SecA molecules. J Mol Biol 340 : 513 24. 174. Angelini, S., Boy, D., Schiltz, E. & Koch, H. G. (2006) Membrane binding of the bacterial signal recognition particle receptor involves two distinct binding sites. J Cell Biol 174 : 715 24. 175. Duong, F. (2003) Binding, activation and dissociation of the dimeric SecA ATPase at the dimeric SecYEG translocase. EMBO J 22 : 4375 84. 176. Kihara, A., Akiyama, Y. & Ito, K. (1995) FtsH is required for proteolytic elimination of uncomplexe d forms of SecY, an essential protein translocase subunit. Proc Natl Acad Sci U S A 92 : 4532 6. 177. van Bloois, E., Dekker, H. L., Froderberg, L., Houben, E. N., Urbanus, M. L., de Koster, C. G., de Gier, J. W. & Luirink, J. (2008) Detection of cross links between FtsH, YidC, HflK/C suggests a linked role for these proteins in quality control upon insertion of bacterial inner membrane proteins. FEBS Lett 582 : 1419 24. 178. Seluanov, A. & Bibi, E. (1997) FtsY, the prokaryotic signal recognition particle recep tor homologue, is essential for biogenesis of membrane proteins. J Biol Chem 272 : 2053 5.

PAGE 205

205 179. Yi, L., Jiang, F., Chen, M., Cain, B., Bolhuis, A. & Dalbey, R. E. (2003) YidC is strictly required for membrane insertion of subunits a and c of the F(1)F(0)ATP synthase and SecE of the SecYEG translocase. Biochemistry 42 : 10537 44. 180. Akiyama, Y., Kihara, A., Tokuda, H. & Ito, K. (1996) FtsH (HflB) is an ATP dependent protease selectively acting on SecY and some other membrane proteins. J Biol Chem 271 : 31196 201 181. Stiburek, L., Fornuskova, D., Wenchich, L., Pejznochova, M., Hansikova, H. & Zeman, J. (2007) Knockdown of human Oxa1l impairs the biogenesis of F1Fo ATP synthase and NADH:ubiquinone oxidoreductase. J Mol Biol 374 : 506 16. 182. Chen, P. M., Chen, Y. Y., Yu, S. L., Sher, S., Lai, C. H. & Chia, J. S. (2010) Role of GlnR in acid mediated repression of genes encoding proteins involved in glutamine and glutamate metabolism in Streptococcus mutans Appl Environ Microbiol 76 : 2478 86. 183. Peng, Y., Luo, Y., Yu, T., Xu, X., Fan, K., Zhao, Y. & Yang, K. (2011) A Blue Native PAGE analysis of membrane protein complexes in Clostridium thermocellum. BMC Microbiol 11 : 22. 184. Abbe, K. & Yamada, T. (1982) Purification and properties of pyruvat e kinase from Streptococcus mutans J Bacteriol 149 : 299 305. 185. Baev, D., England, R. & Kuramitsu, H. K. (1999) Stress induced membrane association of the Streptococcus mutans GTP binding protein, an essential G protein, and investigation of its physiolo gical role by utilizing an antisense RNA strategy. Infect Immun 67 : 4510 6. 186. Wu, J., Cho, M. I. & Kuramitsu, H. K. (1995) Expression, purification, and characterization of a novel G protein, SGP, from Streptococcus mutans Infect Immun 63 : 2516 21. 187. Biswas, S. & Biswas, I. (2005) Role of HtrA in surface protein expression and biofilm formation by Streptococcus mutans Infect Immun 73 : 6923 34. 188. Scott, J. R. & Zahner, D. (2006) Pili with strong attachments: Gram positive bacteria do it differently. Mol Microbiol 62 : 320 30. 189. Huang, M., Meng, L., Fan, M., Hu, P. & Bian, Z. (2008) Effect of biofilm formation on virulence factor secretion via the general secretory pathway in Streptococcus mutans Arch Oral Biol 53 : 1179 85. 190. Pop, O. I., Soprova, Z ., Koningstein, G., Scheffers, D. J., van Ulsen, P., Wickstrom, D., de Gier, J. W. & Luirink, J. (2009) YidC is required for the assembly of the MscL homopentameric pore. FEBS J 276 : 4891 9.

PAGE 206

206 191. Johansen, L., Bryn, K. & Stormer, F. C. (1975) Physiological and biochemical role of the butanediol pathway in Aerobacter (Enterobacter) aerogenes. J Bacteriol 123 : 1124 30. 192. Wilkins, J. C., Homer, K. A. & Beighton, D. (2002) Analysis of Streptococcus mutans proteins modulated by culture under acidic conditions. Appl Environ Microbiol 68 : 2382 90. 193. Yamamoto, Y., Kamio, Y. & Higuchi, M. (1999) Cloning, nucleotide sequence, and disruption of Streptococcus mutans glutathione reductase gene (gor). Biosci Biotechnol Biochem 63 : 1056 62. 194. Herskovits, A. A. & Bibi, E. (2000) Association of Escherichia coli ribosomes with the inner membrane requires the signal recognition particle receptor but is independent of the signal recognition particle. Proc Natl Acad Sci U S A 97 : 4621 6. 195. Maruyama, F., Kobata, M., Kurokaw a, K., Nishida, K., Sakurai, A., Nakano, K., Nomura, R., Kawabata, S., Ooshima, T., Nakai, K., Hattori, M., Hamada, S. & Nakagawa, I. (2009) Comparative genomic analyses of Streptococcus mutans provide insights into chromosomal shuffling and species specif ic content. BMC Genomics 10 : 358. 196. Ling, E., Feldman, G., Portnoi, M., Dagan, R., Overweg, K., Mulholland, F., Chalifa Caspi, V., Wells, J. & Mizrachi Nebenzahl, Y. (2004) Glycolytic enzymes associated with the cell surface of Streptococcus pneumoniae a re antigenic in humans and elicit protective immune responses in the mouse. Clin Exp Immunol 138 : 290 8. 197. Severin, A., Nickbarg, E., Wooters, J., Quazi, S. A., Matsuka, Y. V., Murphy, E., Moutsatsos, I. K., Zagursky, R. J. & Olmsted, S. B. (2007) Proteo mic analysis and identification of Streptococcus pyogenes surface associated proteins. J Bacteriol 189 : 1514 22. 198. Modun, B., Morrissey, J. & Williams, P. (2000) The staphylococcal transferrin receptor: a glycolytic enzyme with novel functions. Trends Mi crobiol 8 : 231 7. 199. Wang, P., Kuhn, A. & Dalbey, R. E. (2010) Global change of gene expression and cell physiology in YidC depleted E. coli. J Bacteriol 200. Soksawatmaekhin, W., Kuraishi, A., Sakata, K., Kashiwagi, K. & Igarashi, K. (2004) Excretion and uptake of cadaverine by CadB and its physiological functions in Escherichia coli Mol Microbiol 51 : 1401 12. 201. Haneburger, I., Eichinger, A., Skerra, A. & Jung, K. (2011) New Insights into the Signaling Mechanism of the Ph Responsive, Membrane Integrated Transcriptional Activator Cadc of Escherichia coli J Biol Chem

PAGE 207

207 202. Griswold, A. R., Jameson Lee, M. & Burne, R. A. (2006) Regulation and physiologic significance of the agmatine deiminase system of Streptococcus mutan s UA159. J Bacteriol 188 : 834 41. 203. Liu, Y., Zeng, L. & Burne, R. A. (2009) AguR is required for induction of the Streptococcus mutans agmatine deiminase system by low pH and agmatine. Appl Environ Microbiol 75 : 2629 37. 204. Liu, Y. & Burne, R. A. (2009) Multiple two component systems of Streptococcus mutans regulate agmatine deiminase gene expression and stress tolerance. J Bacteriol 191 : 7363 6.

PAGE 208

208 BIOGRAPHICAL SKETCH Sara Marie R aser Palmer was born in Jackson Michigan the birthplace of the Re publican Party When she was 11 years old she moved to Somerset Center, Michigan to live with her Aunt and Uncle Carolynn and Terry Palmer, who made their living as artists. She later attended Addison High School in Addison Michigan. With encouragement from her high school b iology teacher, Mr. Ellis, Sara was the first person from her high s choo l to attend a science f air D uring her senior year she won first place in the Tri County, Regional, and State Science Fair s and then went on to compete in the International Science and Technology Science Fair in 2000, in Detroit Michigan where she place d 4 th in her category. After high school, she attended Jackson Community College, where she received an Associate in Science degr ee Sara went on to attend Michigan S tate University and received a B S he moved to Gainesville, Florida in 2005, to pursue a graduate degree in The Interdisciplinary Program in Biomedical Sciences at the University of Flo rida. In her free time she enjoys cooking reading traveling and spending time with her friends and dog Jacopo.