Characterization of the Desiccation Tolerance and Potential for Fomite Spread of Streptococcus pneumoniae
A thesis
submitted by
Rebecca L. Walsh
In partial fulfillment of the requirements for the degree of
Doctor of Philosophy
in
Molecular Microbiology
TUFTS UNIVERSITY
Sackler School of Graduate Biomedical Sciences
February, 2015
Adviser: Andrew Camilli
ABSTRACT
Streptococcus pneumoniae is the causative agent of diseases such as otitis media, pneumonia, and bacterial meningitis. Historically, published literature asserted that pneumococcal transmission occurs via direct contact with respiratory secretions from infected individuals, yet spread of other respiratory tract pathogens has been linked to fomites. As the only known reservoir of S. pneumoniae is the human upper respiratory tract, little work has investigated its capacity to survive in the environment.
Environmental persistence often depends on an ability to withstand desiccation. A knowledge gap exists regarding the viability of vegetative cells, particularly those of major human pathogens, post dehydration. In an era where nosocomial infections are increasingly common, this has become an important area of study.
The goal of this work is to characterize the ability of S. pneumoniae to withstand desiccation. We hypothesize that fomites serve as an alternate source of pneumococcal transmission. To test our hypothesis, we ask three central questions. Can S. pneumoniae survive desiccation? Is S. pneumoniae infectious after desiccation? And which genes or pathways are involved in desiccation tolerance?
Here, we describe the development of a desiccation protocol using plate-grown S. pneumoniae. We show that pneumococci can survive at least 28 days of desiccation under ambient conditions, survival is independent of polysaccharide capsule, and desiccation is a more complicated phenomenon than starvation alone. Multiple serotypes survive desiccation, suggesting that this is a species-wide ability encoded by the core pneumococcal genome. Finally, S. pneumoniae retains infectivity in a mouse model of colonization, lending support to the fomite transmission hypothesis.
ii To explore the pathways correlated with desiccation tolerance, we constructed a transposon insertion library, which was desiccated for two and four days, after which we performed the Tn-seq method, a high throughput sequencing and quantitation of the transposon junctions. By comparing the frequency of reads for each insertion in the pre- and post-desiccation samples, we determined the importance of each gene for S. pneumoniae’s desiccation survival. Of 12 genes selected for further study, mutants in genes encoding the UvrABC nucleotide excision repair complex, SpxB (pyruvate oxidase), SodA (superoxide dismutase), MscS (mechanosensitive channel of small conductance), and cardiolipin synthase all display phenotypes in stress assays. We demonstrate that S. pneumoniae’s desiccation tolerance is tied to its ability to withstand stresses, including those imposed by osmotic changes and reactive oxygen species, leading to a need for DNA damage repair.
iii ACKNOWLEDGEMENTS
I must begin by thanking my advisor, Andrew Camilli, who has been my mentor throughout this challenging, and also very rewarding, journey. He has at times been my teacher, my taskmaster, and my biggest champion. Without his insight, enthusiasm, understanding, and encouragement, I would not be at this point. His love for science is contagious, and a constant reminder to me to be passionate about my own work. He also recognizes the importance of perspective, and is very supportive of and interested in lab members’ non-scientific pursuits. Andy, you have made your lab a wonderful place for my graduate career.
My graduate work has been improved significantly with input from my committee members Ralph Isberg, Carol Kumamoto, and Joan Mecsas. Thank you for your critical evaluations, your helpful suggestions, and your encouragement along the way.
On a daily basis, the Camilli lab members inspired and supported me. In small group discussions, lab meetings, and practice talks, I always received valuable feedback and reassurances. Thank you to all the other S. pneumoniae researchers for your help
(and sometimes your samples) – Tim van Opijnen, Katie Price, Revati Masilamani, Mara
Shainheit, Chelsea Witte, Neil Greene, Ankur Dalia, and Ellie Fleming. Thank you to my various small group members – Ayman Ismail, Bharathi Patimalla, Kim Seed, Mimi
Yen, Heather Kamp, Chelsea, Revati, and Ellie. Thank you to Dave Lazinksi for your suggestions and your crazy bank of knowledge. Thank you to Bharathi for your tireless help with the many mouse experiments. Thank you to lab managers Marc Kimball and
Trish Winterfield for being ordering and organizing whizzes. Thank you to past and present members of the Lion Bay – Evan Bradley, Revati, Chelsea, and Mimi – for
iv making our end of the lab a blast, whether with musical sing-alongs, Disney paraphernalia, venting sessions, or anything to do with our mascot Sir Percival Riggle.
The members of the Molecular Microbiology Program and Department of
Molecular Biology and Microbiology are outstanding; thank you to the students, faculty, post-docs, and staff for making it such a collaborative, nurturing, and fun environment. I also need to thank my classmates – Andrew Hempstead, Julia (Murphy) Keith, and
Jessica Silverman. From studying together first year, to commiserating about research woes, to wrapping up our graduate work, I couldn’t imagine a better group for me to have experienced all of this with. To my close friends that I’ve made during my time here –
Dana, Katie, Marc, Revati, EmilyKate, Stephanie, Julia, Erin, and Mimi – thank you for all of the great memories.
Thank you to Greg Priebe for being a wonderful mentor in my years after college, and for pushing me to go after this. Thanks to friends, near and far, for being so positive and understanding – Shaunagh, Penny, Laura and Tom, Marc and Amy, and many more.
Finally, and most importantly, thank you to my family who have been unwavering in their belief in me. Jill and Paul, thank you for your reminders that I would get through this. Special thanks to Bobbie, who would have been so excited for me. Mom and Dad, thank you for everything; there are no words. My parents instilled a love of learning and science in all of us, and have been my biggest cheerleaders in everything I do. Lisa and
Michael, thank you for your emails, texts, and conversations (scientific and otherwise) that have helped me through this. Gavin, it’s been a long road, and your perspective, confidence, patience, and encouragement continue to amaze me. Thank you for your love and your support.
v TABLE OF CONTENTS
Abstract ...... ii Acknowledgements ...... iv Table of Contents ...... vi List of Tables ...... ix List of Figures ...... x List of Abbreviations ...... xii
CHAPTER 1: Introduction ...... 1 1.1 Streptococcus pneumoniae ...... 2 1.1.1 Capsule and disease ...... 5 1.1.2 Cell membrane and cardiolipin ...... 7 1.2 Bacterial transmission via fomites ...... 8 1.3 Desiccation tolerance ...... 13 1.3.1 Desiccation-tolerant bacteria ...... 13 1.3.2 Effect of relative humidity ...... 14 1.3.3 Effect of season ...... 17 1.3.4 Effect of cell density ...... 18 1.3.5 Preventative responses to desiccation ...... 18 1.3.6 Genes identified with roles in desiccation tolerance ...... 20 1.4 Stress response ...... 23 1.4.1 Oxidative stress response ...... 25 1.4.2 DNA damage repair ...... 28 1.4.3 Competence ...... 31 1.4.4 Protein protection ...... 32 1.4.5 Osmotic stress response ...... 33 1.5 Pneumococcal desiccation tolerance ...... 36
CHAPTER 2: Characterization of the Desiccation Tolerance Phenotype of Streptococcus pneumoniae ...... 38 2.1 Overview ...... 39 2.2 Results ...... 41 2.2.1 Design of a novel desiccation assay ...... 41 2.2.2 S. pneumoniae can survive periods of desiccation of at least 28 days ...... 44 2.2.3 Different growth states support subsequent desiccation survival ...... 46 2.2.4 Desiccation and starvation are separable stresses ...... 49 2.2.5 S. pneumoniae desiccation tolerance is not dependent on the polysaccharide capsule ...... 51 2.2.6 Desiccation tolerance is a property shared by diverse pneumococcal strains ...... 54 2.2.7 Desiccated S. pneumoniae retain the ability to colonize after rehydration ...... 56 2.3 Discussion ...... 59
vi CHAPTER 3: Identification and Physiological Characterization of Genes and Pathways Involved in Streptococcus pneumoniae Desiccation Tolerance ...... 64 3.1 Overview ...... 65 3.2 Results ...... 66 3.2.1 Construction of a transposon insertion library ...... 66 3.2.2 Two- and seven-day desiccation of transposon insertion library ...... 68 3.2.3 Two- and seven-day Tn-seq sample preparation and Illumina sequencing data analysis ...... 73 3.2.4 Four-day desiccation of transposon insertion library and sample preparation for Illumina sequencing ...... 75 3.2.5 Selection of genes involved in desiccation ...... 79 3.2.6 Validation of desiccation changes ...... 81 3.2.7 Stress assays ...... 91 3.2.8 Mfd, UvrA, UvrB, and UvrC are important for surviving UV-induced damage ...... 91 3.2.9 SpxB and SodA are important for surviving hydrogen peroxide stress .....98 3.2.10 MscS, SpxB, SodA, and cardiolipin synthase are affected by osmotic changes ...... 105 3.3 Discussion ...... 115
CHAPTER 4: Summary and Future Perspectives ...... 122 4.1 Summary ...... 123 4.2 Future perspectives ...... 127 4.2.1 Broader experiments ...... 127 4.2.2 Genetic pathways ...... 130 4.2.3 Fomite transmission ...... 132 4.3 Conclusion ...... 134
CHAPTER 5: Materials and Methods ...... 135 5.1 Bacterial strains ...... 136 5.1.1 Strains and growth conditions ...... 136 5.1.2 Strain construction ...... 136 5.2 Desiccation assays ...... 137 5.2.1 Desiccation ...... 137 5.2.2 Mid-exponential-growth-phase desiccation ...... 138 5.2.3 Starvation ...... 138 5.2.4 Competition desiccation ...... 139 5.3 Capsule assays ...... 139 5.3.1 Anti-capsule Western blot ...... 139 5.3.2 Quellung ...... 140 5.4 Mouse model of nasopharyngeal colonization ...... 141 5.5 Library construction and Tn-seq ...... 141 5.5.1 Preparation of genomic DNA ...... 141 5.5.2 In vitro transposition ...... 142 5.5.3 Transformation ...... 143 5.5.4 Making aggregate transposon insertion library ...... 144
vii 5.5.5 Two- and seven-day desiccation of aggregate library ...... 145 5.5.6 Four-day desiccation of aggregate library ...... 146 5.5.7 MmeI method of sample preparation and Illumina sequencing ...... 147 5.5.8 HTML method of sample preparation ...... 148 5.5.9 Data analysis and fitness calculations ...... 149 5.6 Stress assays ...... 151 5.6.1 UV-induced DNA damage ...... 151 5.6.2 Mid-exponential-growth-phase bacteria H2O2 assay ...... 151 5.6.3 Plate-grown bacteria H2O2 assay ...... 152 5.6.4 Growth curve ...... 152 5.6.5 Osmotic shock ...... 153 5.7 Statistical analysis ...... 154
APPENDIX: Fitness Tables ...... 159
REFERENCES ...... 163
viii LIST OF TABLES
Table 3-1 Transposon library complexities ...... 67
Table 3-2 Genes from the 2-day Tn-seq desiccation ...... 74
Table 3-3 Genes from the 4-day Tn-seq desiccation ...... 77
Table 3-4 Genes selected for further study ...... 80
Table 3-5 Compiled results from desiccation and stress assays ...... 113
Table 5-1 Strains used in this study ...... 155
Table 5-2 Primers used in this study ...... 156
Table A-1 DNA repair genes ...... 160
Table A-2 Significant genes identified by Tn-seq found in D39, but not
TIGR4 ...... 161
Table A-3 Significant genes identified by desiccation Tn-seq, but unresponsive
in other Tn-seq conditions ...... 162
ix LIST OF FIGURES
Figure 1-1 Pathogenic route of invasive pneumococcal disease ...... 4
Figure 1-2 Pneumococcal production of reactive oxygen species ...... 27
Figure 2-1 Schematic of desiccation assay ...... 43
Figure 2-2 S. pneumoniae D39 survival after desiccation ...... 45
Figure 2-3 Schematic of exponential growth-phase desiccation assay ...... 47
Figure 2-4 Plate-grown versus exponential growth-phase cell desiccation ...... 48
Figure 2-5 S. pneumoniae D39 survival after desiccation versus nutrient
deprivation ...... 50
Figure 2-6 Desiccation tolerance of S. pneumoniae D39 (encapsulated) and its
acapsular derivative AC326 ...... 52
Figure 2-7 Anti-capsule Western blot of desiccated S. pneumoniae ...... 53
Figure 2-8 Desiccation tolerance of 17 S. pneumoniae strains ...... 55
Figure 2-9 Schematic of murine nasopharyngeal colonization protocol ...... 57
Figure 2-10 Murine nasopharyngeal colonization by desiccated versus
non-desiccated S. pneumoniae ...... 58
Figure 3-1 Schematic of desiccation of transposon insertion library ...... 70
Figure 3-2 Gel of total-cell and live-cell gDNA from T0, T2, and T7 library
samples ...... 72
Figure 3-3 Bacterial strain construction ...... 82
Figure 3-4 WT versus ∆spd_0563 4-day competition desiccation ...... 84
Figure 3-5 WT and mutant survival after 4-day single-strain desiccation ...... 85
Figure 3-6 WT versus mutant 4-day competition desiccation ...... 87
x Figure 3-7 WT and ∆spxB desiccation survival ...... 89
Figure 3-8 Schematic of UV stress assay ...... 93
Figure 3-9 WT survival after UV treatment ...... 94
Figure 3-10 Bacterial survival after UV treatment ...... 96
Figure 3-11 Schematic of H2O2 stress assay ...... 99
Figure 3-12 WT survival after exposure to varying levels of H2O2 ...... 100
Figure 3-13 Bacterial survival after H2O2 treatment ...... 102
Figure 3-14 Growth curve of WT and ∆spxB with and without Oxyrase ...... 104
Figure 3-15 Schematic of osmotic downshock assay ...... 108
Figure 3-16 Bacterial survival after osmotic downshock ...... 109
Figure 3-17 Bacterial survival after osmotic upshock ...... 112
Figure 4-1 Pneumococcal transmission routes ...... 126
xi LIST OF ABBREVIATIONS
BER: Base excision repair cat: Chloramphenicol acetyltransferase
CDC: Centers for Disease Control
CI: Competitive index
CFU: Colony forming units
Cm: Chloramphenicol
CO2: Carbon dioxide
CSP: Competence-stimulating peptide dC: Deoxycytosine dH2O: Distilled deionized water dsDNA: Double-stranded DNA
ECM: Extracellular matrix
EtOH: Ethanol gDNA: Genomic DNA
Gd3+: Gadolinium ion
Gent: Gentamicin
H2O2: Hydrogen peroxide
HSP: Heat-shock protein
HTML: Homopolymer tail-mediated ligation
ICU: Intensive care unit
IPD: Invasive pneumococcal disease
K+: Potassium ion
xii LPS: Lipopolysaccharide
MIC: Minimum inhibitory concentration
MMR: Mismatch repair
Mn2+: Manganese ion
MRSA: Methicillin-resistant Staphylococcus aureus
NER: Nucleotide excision repair
O2: Oxygen
OD600: Optical density at 600 nm
PBS: Phosphate-buffered saline
PCR: Polymerase chain reaction
PE: Phosphatidylethanolamine
PFGE: Pulsed-field gel electrophoresis
PG: Phosphatidylglycerol
RBS: Ribosome binding site
RH: Relative humidity
SDS: Sodium dodecyl sulfate
SEM: Standard error of the mean
SOE: Splicing by overlap extension
Spec: Spectinomycin
Spp: Species ssDNA: Single-stranded DNA
SN: Supernatant
T0: Initial time point
xiii TBS: Tris-buffered saline
TCS: Two-component system
THY: Todd Hewitt broth supplemented with yeast extract
UV: Ultraviolet
VBNC: Viable-but-non-culturable
WHO: World Health Organization
WT: Wild-type
xiv CHAPTER 1: Introduction
1 1.1 Streptococcus pneumoniae
Streptococcus pneumoniae (the pneumococcus) is a Gram-positive diplococcus and common member of the human nasopharyngeal microbiota. In fact, the only known natural niche of S. pneumoniae is the human upper respiratory tract (26, 239, 249). At any given time, 5-10% of healthy adults are asymptomatically colonized with S. pneumoniae (26), whereas on average, 55% of 3-year-olds are pneumooccal carriers (19).
Current medical dogma asserts that transmission occurs through direct contact with respiratory secretions from infected individuals (229). Everyday actions such as coughing, sneezing, and nose-blowing spread these aerosolized respiratory droplets (74,
235, 250), enabling colonization of a new host. With a break in the host immune defenses, the pneumococcus can switch from commensal to pathogen and cause invasive disease. By definition, invasive pneumococcal disease (IPD) occurs when the bacteria access normally sterile spaces such as the inner ear, lung, and blood, causing diseases including otitis media, pneumonia, and bacterial meningitis, respectively (155, 248).
Based on polysaccharide capsular differences, more than 90 serotypes of S. pneumoniae have been identified. Even with the advent of several highly effective multivalent capsular polysaccharide-protein conjugate vaccines, S. pneumoniae continues to be a devastating pathogen worldwide. Pneumococcal diseases remain a major medical and financial burden in developed countries, while still causing millions of deaths every year, with the very young and the elderly comprising the most susceptible populations
(19, 26, 50, 165, 246-248). Although ear infections are a relatively mild form of IPD, they are nonetheless the primary reason, other than wellness visits, for children in the
United States to visit the doctor (26, 157). Annually, six million ear infection cases are
2 caused by the pneumococcus, generating a treatment industry worth two billion dollars in this country alone (157, 247). In the United States and Europe, S. pneumoniae is the leading cause of community-acquired bacterial pneumonia (247, 248). In developing countries, the situation is even direr. Data from the last decade indicate that over one million children under the age of five die every year as a result of pneumococcal acute respiratory infections (247). In total, the World Health Organization (WHO) attributes
1.6 million deaths annually around the world to S. pneumoniae.
3
!"#$%#&'()#%*+',-(
!-*"#13%&( 61-%*71#8&9'1+(51##"19'( 4%51+(-*#'1)(
!+2'%+"( A&&'#('1#( ;"&/-'-( .&'/0%&"1( <33-(0')"1( ;"&/-"3-(
.+'/#1( .'#"51#)"/0( :+%%)( =0*8'01( =0*8'01( ;'*35'0"1(
.'#",%&'/0( ?%"&,-( >'&"&9'-( .'#",%&"3-( !#,7#"3-@( >'&"&9"3-( <-,'%08'+"3-(
Figure 1-1: Pathogenic route of invasive pneumococcal disease. Adapted from references (19, 165).
4 1.1.1 Capsule and disease
Pneumococcal disease is believed to begin with asymptomatic colonization (Fig.
1-1), establishing what is termed the carrier state (19, 103, 120, 155, 239) in which the bacteria are mainly located in nasopharyngeal biofilms (136, 154). A major virulence factor of the pneumococcus is its polysaccharide capsule, which is useful for the initial steps of pneumococcal colonization of the nasopharynx and critical for later immune evasion (2, 19, 103). Acapsular pneumococcal variants have been shown to have significantly reduced virulence in mice (2), and accordingly were thought to be avirulent.
Acapsular bacteria can colonize, albeit less efficiently than encapsulated isotypes (133,
159). The negatively-charged capsule is believed to enhance bacterial movement through the nasal mucus, thereby allowing closer proximity to the epithelial surface (103, 159).
Once in contact with the epithelium, the capsule is disadvantageous and so the bacteria down-regulate capsule expression, exposing surface structures that act as adhesins and are therefore necessary for binding to host cells (19, 103). In systemic infection, capsule is up-regulated, as its expression protects against phagocytosis by obscuring the bacterial surface ligands that could be recognized and targeted by opsonins such as complement and by professional phagoctyes (21, 75).
Capsule expression is tightly regulated, and the cps operon contains genes encoding enzymes for serotype-specific synthesis, polymerization, and export of the sugar chains (205, 254). Regulation may occur post-translationally via protein phosphorylation and it also may occur transcriptionally at the promoter region, leading to phase variation, a reversible genetic change between two distinct capsular states (109,
240). High levels of capsule, and consequently enhanced virulence in the blood due to
5 resistance to opsonophagocytosis, characterize opaque colonies. Transparent colonies display low levels of capsule, making these variants better at epithelial cell binding and nasopharyngeal colonization (19, 109).
While the capsule is not static within the host, neither is it fixed between pneumococcal strains. The capsular polysaccharides are highly diverse, with the different structural and immunogenic combinations differentiating the more than 90 known serotypes (19, 239). The capsule identifies serotypes, but it is not the only variable characteristic among pneumococci. S. pneumoniae has high genome plasticity, such that the genetic content between strains can vary by up to 10% (71, 78, 119, 229).
With this genetic variation, there are also differences in virulence. Although most serotypes have the potential to cause severe disease, only certain serotypes account for the overwhelming amount of pneumococcal diseases globally (26, 247). The Centers for
Disease Control (CDC) report that the ten most prevalent global serotypes cause 62% of
IPD (26). Because specific serotypes account for most pneumococcal disease, vaccine strategy has focused on targeting the immune response against the capsular polysaccharides of these ‘more relevant’ serotypes. Two types of vaccines are currently in use. The 23-valent polysaccharide vaccine is made of purified pneumococcal polysaccharides from the 23 serotypes that caused 88% of bacteremic disease at the time of its inception (26, 247). A 10-valent and a 13-valent conjugate vaccine are also available (248). The 13-valent conjugate vaccine targets seven serotypes that caused the majority of disease in the United States in the late 1990’s, and an additional six that emerged as newly pathogenic vaccine escape strains after the introduction of a 7-valent vaccine in 2000 (26).
6 One side effect of large-scale vaccination programs is that the selective pressure against vaccine serotypes allows for an increase in non-vaccine serotypes that colonize, and therefore potentially cause disease (19). Another consequence has been the emergence of vaccine escape strains, which are strains that have acquired a new capsular locus, and thus serotype, by horizontal gene exchange (26, 35, 36, 100). Such horizontal genetic exchange is aided by the natural transformation ability of the pneumococcus.
1.1.2 Cell membrane and cardiolipin
Other vaccine efforts have concentrated on surface proteins, which are anchored in the cell wall and the cell membrane. S. pneumoniae’s plasma membrane contains mainly phospholipids, glycolipids, and proteins (8). It is comprised mostly of two glycolipids –monoglucosyldiacylglycerol and galactosylglucosyldiacylglycerol – and two polar lipids – phosphatidylglycerol (PG) and cardiolipin (60, 61, 175). To make cardiolipin, the phospholipase cardiolipin synthase catalyzes the transfer of a phosphatidyl moiety from one PG to another in a process known as transphosphotidylation (190, 199), yielding cardiolipin and glycerol. It is believed that cardiolipin synthase also may act in the reverse direction (199). Escherichia coli cells lacking cardiolipin synthase have trace amounts of cardiolipin, suggesting there are other enzymes involved in its synthesis, however there are conflicting reports regarding the presence of cardiolipin in similar Bacillus subtilis mutants (190). Cardiolipin was found to accumulate as bacterial growth slowed (25, 199), and therefore is present more in stationary than in exponential phase streptococci. S. pneumoniae lacking cardiolipin synthase has a reduced ability to both form biofilms and to colonize mice (154).
7 Cardiolipin’s function in membranes is several-fold. As a dimeric lipid (two PG groups linked by one glycerol), it can serve as a flexible linker to fill cavities at protein interfaces, thus stabilizing both intra- (subunit interactions within individual proteins) and inter-protein (between proteins in complex) interactions (146). While cardiolipin can be heterogeneously distributed around the membrane, it also participates in establishing membrane curvature by forming lipid-protein domains, and so primarily localizes to bacterial poles and to the septa during E. coli and B. subtilis division (105, 143, 146).
This makes sense, as PG synthase and cardiolipin synthase, too, showed septal localization dependent on division protein FtsZ (localizes at the septum to organize cellular division machinery) (146, 161). Because aggregates of cardiolipin localize to the cell poles, it has been suggested that they may recruit and bind cell cycle and division proteins (146). This hypothesis is supported by microscopy work showing that E. coli cells deficient in phosphatidylethanolamine (PE), and therefore with elevated PG and cardiolipin levels, have cell division proteins that aggregate at cardiolipin domains, thereby inhibiting division (147). Clearly, cardiolipin is very important for normal cell function because of its involvement in both membrane protein organization and bacterial division.
1.2 Bacterial transmission via fomites
Commonplace activities including talking, coughing, and sneezing spread large amounts of bacteria into the external environment (74, 235, 250), and research on respiratory tract pathogens has implicated dust and disintegrating sputum as reservoirs of bacterial transmission (236). Fomites, or environmental surfaces, have been described as
8 probable or confirmed sources of infections caused by a variety of human pathogens (1,
96, 102, 148, 158, 162, 197, 202, 221). The study of bacterial persistence on fomites is increasingly important in the modern age of nosocomial infections, many of which may be linked to transmission via fomites.
Acinetobacter baumanii is a Gram-negative environmental bacterium that can cause opportunistic infections, especially among the immunocompromised and those in the intensive care unit (ICU). When desiccated on glass cover slips under ambient conditions, 39 strains of A. baumanii were shown to survive for an average of 22 to 33 days, and this desiccation tolerance was postulated to contribute to the bacteria’s ability to survive and cause outbreaks (96). Another common nosocomial pathogen, the Gram- positive Staphylococcus aureus has become an even greater threat due to an increase in methicillin-resistant strains (MRSA). Survival studies on both nosocomial and community-acquired strains desiccated in combination with sterile blood or pus showed
MRSA survived on the surface of coins for up to 13 days (221). In addition, hospital television sets, furniture, electrical equipment, and surgical ward surfaces have been proven to harbor MRSA, indicating that these surfaces may act as bacterial reservoirs for nosocomial transmission (18, 139, 211). Corynebacterium diphtheriae, the Gram- positive causative agent of both cutaneous and respiratory diphtheria, is transmitted via direct person-to-person contact or through respiratory secretions (241). It, too, has been shown to survive for extended periods on a variety of fomites. Depending on the surface,
C. diphtheriae was recoverable for up to 180 days after initial desiccation (236).
Fewer experiments have investigated the ability of bacteria to maintain infectivity post-desiccation, although the number of such studies has increased of late. In one
9 example, Streptococcus pyogenes-contaminated human blood was dried on paper towels for 20 weeks. Of the strains tested, 80% survived and 70% retained the ability to grow in fresh blood, which the authors took as evidence of maintained virulence (184). The ability of S. pyogenes to survive and retain infectivity during desiccation could prove particularly dangerous for several reasons: 1) cases of nosocomial transmission have been postulated to result from contaminated fomites (102, 197), and 2) this pathogen has a high mortality rate in patients with severe infection (24).
Epidemiologically, it still is rare that infections are attributed to fomite transmission, but outbreaks in military and care facilities have been associated with contaminated fomites. A 1946 study attributed pneumococcal spread between two military squadrons to fomites, arguing: 1) the two groups had no contact and therefore respiratory droplet spread was not possible, 2) serotype distribution between squadrons was very similar, and 3) cultures of up to 39% of floor dust contained viable pneumococci (83). More recently, a group at Tripler Army Medical Center in Honolulu used pulsed-field gel electrophoresis (PFGE) to identify the same strain of MRSA in two patients, on the computer keyboards and faucet handles in their rooms, and on keyboards elsewhere in the unit (22). In outbreaks of community-acquired MRSA, infection was associated with exposure to contaminated fomites including whirlpools and benches, as well as shared razors, soaps, and towels (148, 160). The only location from which S. aureus was cultured during an outbreak in a French ICU was the finger SpO2 sensor, and
PFGE was employed to establish that the same epidemic strain infected multiple patients whose only possible source of contamination was these sensors (170). In a nursing home outbreak of streptococcal infections, S. pyogenes was found in the carpets, chairs, and
10 curtains of the contaminated facility, but not in those from a separate nursing home (197).
Five patients and one staff member tested positive for S. pyogenes infection during the outbreak, but no additional infections occurred after rigorous environmental cleaning measures were implemented. In a 21-month-long outbreak in a Florida hospital burn unit, 43 of 103 patients admitted contracted an Acinetobacter calcoaceticus infection, and an additional 20 become asymptomatically colonized (206). Mattresses were the lone environmental surface tested with a significant frequency of Acinetobacter isolation, and only a new directive to dispose of mattresses after patient discharge led to the complete eradication of A. calcoaceticus from the burn unit. In a three-month-long nosocomial outbreak of multidrug-resistant Pseudomonas aeruginosa in a pediatric oncology unit, a single strain, as identified by PFGE, infected eight patients (23). The only fomites from which this same strain was isolated were shared bath toys; after removal of these toys, no subsequent infections occurred. Although none of these fomites have been proven conclusively to be the source of transmission, the evidence strongly supports that hypothesis.
Environmental sampling studies have found that fomites harbor many bacterial and viral species, and therefore may serve as vectors of disease transmission. Perhaps most troubling is the possibility that fomites found in healthcare settings, where many patients are already immunocompromised, can transmit infectious agents. Swabbing of stethoscopes, reusable tourniquets, x-ray cassettes, pens, hand sanitizer dispensers, doorknobs, faucets, and hospital curtains has shown that all are contaminated with potentially dangerous microbes (22, 47, 73, 110, 134, 159, 178, 238, 244, 255). In addition to surfaces regularly contacted by patients, medical staff members exclusively
11 use hospital computers, medical charts, and their personal mobile phones and white coats
– all potential bacterial reservoirs that could enable transfer from fomite to clinician to patient (217). A survey of plastic medical charts in the general wards and specialized units of a Taiwan hospital found that 63.5% and 83.2%, respectively, were contaminated with culturable bacteria (28). Other than the Obstetrics ICU, charts from the Medical,
Surgical, and Pediatric ICUs had significantly higher contamination rates, possibly because of the invasive procedures that occur in these units and the general ill health of the patients. Numerous other studies have also found medical charts to be highly contaminated, and many of the bacterial strains are common causes of drug-resistant infections (3, 13, 76, 219, 256). Mobile phones are widely used by medical professionals in the hospital environment, and multiple studies have shown bacterial contamination of said phones. The hands and phones of 200 medical workers, including doctors, nurses, and support staff, at a hospital in Turkey were tested by wiping with a moistened sterile swab (228). After streaking the swabs onto selective agar, colony growth was seen from
94.5% of the phones, totaling 307 separate species of bacteria, molds, and yeasts (12 of which were streptococci). From the hands of these staff, 333 separate species were identified, many of which overlapped with those found on their phones, suggesting transmission between hand and phone (228). In a separate study in Nigeria, swabs of
62% of mobile phones tested were positively cultured for pathogenic bacterial species
(1). The rate of bacterial carriage on healthcare workers’ phones was over 42%; in contrast, 92% of food vendors’ phones were contaminated. These studies are especially relevant, as the bacteria were culturable from the swabs onto plates or in media, proving they remain alive on the fomite. Some studies sample bacterial diversity via 16S rRNA
12 sequencing, prepared from DNA acquired on surface swabs (123). It could be contended that such DNA comes from dead cells and therefore does not represent the live bacterial population on the fomite, however colony growth eliminates this argument. Finally, several studies have shown controlled, in vitro transfer of nosocomial pathogens from fomites to healthcare workers’ hands in the absence of any patient contact, a most compelling argument for fomite transfer of microbes in the hospital setting (169).
1.3 Desiccation tolerance
1.3.1 Desiccation-tolerant bacteria
Survival on fomites often necessitates an ability to survive periods of desiccation.
Antonie van Leeuwenhoek was the first to document “animalcule” recovery after desiccation (37, 106, 180). Now, more than 300 years later, the archetypal microbes capable of surviving desiccating conditions are the spore-formers, cyanobacteria, and extremophiles (16, 17, 179, 180, 204). Spore-forming bacteria such as B. subtilis respond to nutrient limitation and other stresses by inducing a signal cascade that results in sporulation. The mother cell replicates its chromosome and septates, packaging the DNA into the newly forming spore. Conditions of the spore, including numerous low- permeability layers, cessation of metabolism, and decreased water content, allow for the protection of DNA and proteins from extracellular and intracellular damaging agents
(203). Spores can remain in this dormant state for prolonged periods, even withstanding extreme dryness, heat, oxidizing agents, and radiation, until external conditions signal a start to the germination cascade (203). Among non-spore-formers, these same stresses are also endured by the extremophile Deinococcus radiodurans, which is recognized for
13 its ability to tolerate ionizing radiation, extreme temperatures, and desiccation (209).
Less is known about the environmental survival of other vegetative cells, especially in regards to their desiccation tolerance. Now that it is apparent that a variety of human pathogens are capable of persisting on fomites, sites where they experience stresses including desiccation, more work is being devoted to this area of study.
1.3.2 Effect of relative humidity
An assortment of factors play into desiccation tolerance, a number of which are the result of chance circumstances more than bacterial preparedness, and the roles of all remain unclear. There still is not a consensus in the field as to the effect of atmospheric relative humidity (RH) on bacterial survival. Dunklin and Puck argued for the existence of an unfavorable cellular water content (present at intermediate, but not low or high, humidities) that continues to allow cellular processes to function after drying, leading to damage caused by toxic products of metabolism and respiration (45). In experiments on aerosolized S. pneumoniae, they determined that the bacteria displayed high mortality at
50% RH (no recoverable pneumococci by 10 minutes), but survived at low or high humidities (~35% survival at 20% RH and ~16% survival at 80% RH after two hours).
They noted two decay rates: 1) a rapid loss in viability in the first 5-20 minutes after aerosolization, and 2) a much slower rate of loss over the remainder of the experiment
(45). The rapid decay was most pronounced at 50% RH, and least extreme at 20% RH, suggesting that low moisture in the air is most conducive to pneumococcal survival. The other two bacteria tested, group C streptococcus and Staphylococcus albus, also lost viability most rapidly at 50% RH. Across 28 days of desiccation, D. radiodurans survival had an inverse relationship with RH, with nearly 100% survival in an ultrahigh
14 vacuum, ~12% survival at 4% RH, ~10% survival at 33% RH, and 0.03% survival of cells suspended in PBS (12). Aerosolized Francisella tularensis disseminated from the wet state (culture) were not viable after 15 minutes at 55-60% RH, though when disseminated from the dry state (freeze-dried powder), survival correlated with lower RH
(34). Dry-disseminated E. coli, too, showed highest recovery at low RH (33), indicating that low environmental moisture likely allows the bacteria to remain in a desiccated state.
Transmission of aerosolized respiratory viruses also is more efficient at lower humidity
(35% RH) than at high (80% RH) (132). At all humidity levels, bacteria in larger aerosolized droplets died quicker than bacteria in small droplets (45). This could be due to larger droplets simply containing more injurious agents (e.g. salts in the suspending medium), which would be concentrated to damaging levels upon dehydration, or it could be because larger droplets require more time for full evaporation, allowing the cells to stay metabolically active longer and thereby accumulate greater amounts of toxic byproducts.
Kramer and colleagues surveyed numerous published reports and concluded that most nosocomial pathogens investigated, other than S. aureus, survived on fomites best at high humidities (113). One such study found significantly higher recoverability of
Acinetobacter spp. at 93% and 31% RH (11-day survival) compared to those kept at 10%
RH (4-day survival) (95). Chlamydia trachomatis survived on a plastic surface for several hours under 100% RH, versus a 30-minute survival at 56% RH (162), however it could be argued that such high humidity precludes full desiccation. Shigatoxin- producing E. coli survived best at 98% humidity (152) probably because, although the bacterial droplets dried within one hour, water droplets soon condensed on the fomite and
15 therefore could have allowed bacterial hydration during “desiccation.” Alternatively, other researchers proposed that the slow drying process that occurs at high humidities causes more distortion of protein structures and cell membranes than the fast drying that takes place at low RH (193). There is yet another camp that argues that RH has a minimal effect and that other variables have significantly greater influence on desiccation tolerance. When Cox observed survival of aerosolized F. tularensis at low and high humidities, he resolved that survival after drying is not directly related to water content
(34), although this explanation does not account for its poor viability at mid-range RH.
From comprehensive evaluations of the literature, multiple groups ultimately concluded that the effect of humidity on environmental survival varies widely among pathogens
(112, 218, 236).
Few published studies report the water content of desiccated bacteria. It is known that desiccation-sensitive bacteria die if their cellular water content is reduced to 0.3 grams (g) of free water per gram of dry weight (30% residual moisture) (16), whereas anhydrobiotes (organisms whose metabolism stops upon desiccation) have water contents below 10% (54). Therefore, 10% or less residual moisture is considered true desiccation
(54, 209), although most organisms never reach this level of dehydration. Among spore- forming bacteria, which are not considered anhydrobiotes, spore coats have water contents of 0.3-0.7 g water per g dry weight, while growing cells are measured at 3-4 g water per g dry weight (203). Reduced water content is associated with increased DNA damage and decreased bacterial viability. For example, in E. coli, water content below
15% is closely correlated to an increase in the number of single-strand DNA breaks (9).
16 1.3.3 Effect of season
Humidity, and thus cellular water content, is often decided by seasonal weather patterns. Certain microbes cause more infections during the summer months, while others are known to cause disease mostly in winter. Rates of Acinetobacter spp. isolation from hospital fomites, as well as rates of nosocomial infection, increase during the warm weather, a trend that Jawad and colleagues attributed to increased RH and elevated sweating which may favor colonization (95). Lower temperatures, though, are almost universally associated with increased pathogen survival (7, 111, 113, 218). Currently, no direct causality has been shown between a seasonal characteristic and increased pathogen survival and transmission. Humidity, precipitation, temperature, wind patterns, ozone levels, air pollutant levels, human behavior, and more may all play a role.
Rates of pneumococcal carriage and IPD vary seasonally, peaking in the winter months just like viral respiratory infections (44, 63, 111, 243). In a three-year study, the seasonal peak of IPD correlated with the number of hours of darkness, but not with total seasonal precipitation (44). The authors argue that similar seasonal patterns in IPD seen at multiple geographic locations, which experience a range of temperatures and precipitation, indicate that the one constant is decreased daylight. Whether this results in circadian changes in host immune systems, reduced vitamin D production and thus reduced immunity, prolonged indoor contact among people (due to cold or darkness) increasing the chances of pneumococcal transmission, or elevated environmental survival of the bacteria due to reduced ultraviolet (UV) exposure remains to be seen. A separate study found an inverse correlation between IPD and ozone levels; ozone is related to the amount of solar rays (111), i.e. UV light, indicating that low levels of UV allow greater
17 pneumococcal environmental survival (in respiratory droplets or on fomites). Finally, evaluating multiple variables on a weekly time-scale led another group to conclude that increases in UV radiation are most strongly associated with decreased IPD in
Philadelphia (243).
1.3.4 Effect of cell density
Another factor that has been considered is bacterial cell density at time of desiccation. Most research suggests that high cell density increases overall bacterial recovery (27, 113, 158, 179, 237). This may be due to a phenomenon known as cryptic growth, in which living cells recycle nutrients from dead cells in order to survive (237).
Cryptic growth, however, has been associated exclusively with cells under starvation conditions, only one of the likely many processes occurring during desiccation. On the other hand, the survival kinetics of both S. aureus (117) and Listeria monocytogenes were independent of starting concentration, leading the authors to conclude that desiccation tolerance is a function inherent to the cells and independent of proximity to neighboring bacteria (82). As seen with S. pneumoniae (45), there were two rates of decay in L. monocytogenes survival curves (82), a trend also observed in multiple Salmonella species
(51, 94, 126, 151), S. aureus (27), A. baumanii (58), and Neisseria meningitidis (227).
1.3.5 Preventative responses to desiccation
Certain characteristics, many of which are shared by these nosocomial pathogens, have been shown to contribute to microbial ability to resist periods of desiccation. Cocci are believed to survive drying at higher rates than rod-shaped cells because a round shape and low surface area-to-volume ratio are adaptations to desiccating conditions (186).
Overall, Gram-positive bacteria seem to tolerate desiccation better and for far longer than
18 their Gram-negative counterparts (54, 69, 94, 150, 152). This likely is due in part to their differences in cellular architecture; in particular, the much thicker peptidoglycan layer of
Gram-positive cell walls are less likely to rupture during desiccation (150, 172). A major component of the outside of many bacteria is a layer of polysaccharide, for example as a capsule or part of the lipopolysaccharide (LPS) molecule attached to the outer membrane of Gram-negatives. Mucoid strains of E. coli, A. calcoaceticus, and Erwinia stewartii displayed significantly increased resistance to desiccation compared to isogenic non- mucoid mutants (168). Furthermore, using a β-galactosidase fusion, the authors demonstrated desiccation-induced transcription of the capsule operon. Salmonella enterica serovar Typhimurium LPS mutants were used to validate protective effects of polysaccharide on survival after drying for 24 hours (56). It has been suggested that extracellular polysaccharides can trap water molecules close to the cell surface, thus increasing bacterial survival under desiccating conditions (179). Some research has even shown that bacteria, including Pseudomonas spp., dramatically increase polysaccharide production in response to desiccation, such that they contain more exopolysaccharide than total protein when desiccated (189). A study on the survival of freeze-dried bacteria, however, found an inverse relationship between cell surface polysaccharide presence and bacterial survival. The authors suggest that too many linked sugars on the outside of the cell, in the form of both polysaccharides and LPS, would retain more water molecules, which may allow for short-term survival but inhibit long-term desiccation tolerance
(150). A caveat of this work is that they solely investigated freeze-drying survival, and the presence of water during bacterial freezing can create ice crystals, which are known to be damaging to cell structures (92).
19 To protect against damage like ice crystals, bacteria will accumulate protective sugars when faced with freezing or desiccating conditions. This is because water is crucial for many cellular activities and with its loss, essential processes such as protein folding and protein-protein interactions are disrupted (195). Water acts as a nucleophile in certain reactions, helps maintain or change concentrations of various molecules via participating in diffusion, osmosis, and transport, and acts as the driving force for hydrophobic interactions, such as those involved in protein folding, DNA structure, and lipid bilayer formation. Therefore, a compensatory mechanism that some plants and microbes utilize is the accumulation of disaccharides, specifically trehalose, sucrose, maltose, and cellobiose, which act as osmoprotectants by maintaining the organization of membranes and replacing water in molecular interactions (37, 92, 108, 150, 179). This both allows cellular structures to remain intact and prevents the denaturation and damage of proteins (92, 179). In fact, experiments on desiccation-sensitive recombinant E. coli have demonstrated that engineered expression of sucrose and trehalose can increase recovery after drying by up to 10,000-fold (17, 108). Other work has shown that trehalose-deficient mutants of Saccharomyces cerevisiae and Candida albicans are extremely sensitive to oxidative stress, suggesting that trehalose acts as a protectant against free radical damage in various systems (92).
1.3.6 Genes identified with roles in desiccation tolerance
Studies on desiccation tolerance in various bacteria have identified a number of genes with roles in surviving this process. S. aureus alkyl hydroperoxide reductase
(AhpC) and catalase (KatA), which have linked antioxidant functions, are needed for wild-type (WT) levels of survival after a 24-hour desiccation period (32). Single
20 deletions for ahpC and katA show 10% and 50% survival relative to WT, respectively, and a double mutant survives at only 1% of the WT level (32). By screening both a
Mariner transposon insertion library and defined mutants, Chaibenjawong and Foster identified clpX, sigB, and yjbH as necessary for S. aureus desiccation tolerance (27).
ClpX is part of the ClpXP protease, which is involved in protein turnover, and is believed to have a regulatory role in the S. aureus osmotic and oxidative stress responses (53).
YjbH is hypothesized to function as an adapter protein by binding and delivering Spx
(suppressor of clpP and clpX) to the ClpXP protease for degradation (27, 55). B. subtilis
Spx acts as a redox sensor and up-regulates oxidative stress response genes (156), and may behave similarly in additional bacterial species. S. aureus SigB is an alternative sigma factor that controls over 250 genes and has a role in the stationary phase stress response and the oxidative stress response (27). The authors suggest that SigB is related to desiccation tolerance via its ability to up-regulate genes involved in the response to oxidative stress (27). E. coli alternative sigma factor RpoS, which is involved in the stress response during stationary phase and regulates the synthesis of osmoprotectant trehalose, is critical for desiccation tolerance, as mutants displayed 1,000,000-fold reduced survival over four days of desiccation versus WT (212).
In other bacteria, certain metabolic proteins and osmoprotectant transporters were identified. A study on two Salmonella strains found that genes involved in fatty acid metabolism, transport, osmotic resistance, cell envelope functions, the stress response, and more were differentially regulated after two hours of desiccation (126). A four-hour desiccation of S. enterica serovar Typhimurium led to significant up-regulation of osmoprotectant transporters proP, proU, and osmU (51). Individual deletion mutants
21 displayed reduced survival on a stainless steel surface, although ∆proP showed the most significant survival defect. Yet another study on this bacterium using microarrays found
90 genes were expressed and seven were down-regulated after 22 hours of desiccation
(65). Desiccation induced expression of ribosomal structural genes, as well as genes involved in amino acid metabolism, ion transport, and stress response; the highest induction was seen in the kdpFABC operon, which codes for a potassium transport channel (65). Genes encoding Salmonella isocitrate-lyase AceA, lipid A biosynthesis protein Ddg, iron-sulfur cluster scaffolding protein NifU, global regulator Fnr, and alternative sigma factor RpoE are all important for surviving desiccation, and all of which have also been shown to be involved in bacterial survival during oxidative stress, as well as heat and cold shock (65). Some of these proteins, however, were important only for long-term survival (12 weeks) during cold desiccation, indicating that there may be different groups of genes induced depending on differing conditions of desiccation. Two separate proteomics approaches identified hundreds of proteins involved in A. baumanii persistence during desiccation (58). These proteins are in a variety of pathways, with functions of antimicrobial resistance, oxidative and nitric stress response, chaperones, metabolism, transport, and more. Select up-regulated proteins include superoxide dismutase, catalase, DNA repair protein UvrB, and chaperone ClpB (58). By and large, more transcription, translation, and metabolism proteins were down-regulated than up- regulated, implying that these processes are largely turned off during desiccation and that the cells may survive in a dormant state (58). A transposon mutagenesis screen identified regulatory genes relA, rpoE2, and hpr and DNA repair gene uvrC as important for soil bacterium Sinorhizobium meliloti to withstand desiccation (87). Subsequent desiccation
22 assays on individual uvrA, uvrB, and uvrC deletion mutants showed them to have 10-fold more killing after four weeks and 100-fold more killing after eight weeks of desiccation than WT, confirming the role of DNA repair in desiccation resistance (87).
1.4 Stress response
Bacteria frequently experience stressful conditions, which may include heat, UV exposure, nutrient limitation, osmotic, matric (low water content), pH, or oxidative stress.
In response, specific genes or general pathways are induced to react to potential damage and increase the chances of survival. Exposure to one stress is known to cause bacteria to become more tolerant to other types of stresses (66, 114). In fact, desiccated Salmonella spp. demonstrated increased survival as compared to non-desiccated cells under stresses such as high heat, salts, hydrogen peroxide (H2O2), and UV irradiation (66). Many stress and starvation signals lead to the production of a core set of stress response proteins; this stress regulon is controlled by alternative sigma factor σB in Gram-positives such as B. subtilis, and sigma factor RpoS in Gram-negatives like E. coli (11, 194, 201). The B. subtilis σB-dependent regulon is composed of approximately 150 genes that encode a variety of proteins, such as catalases, transporters, and osmoprotectant synthases (201).
Activation of σB results from either glucose starvation or physical stresses like extreme temperature or salt stress (201). Specific stress signals induce other sigma factors. For example, oxygen stress induces B. subtilis σA, leading to expression of antioxidants and catalases, and E. coli RpoD (σ70), leading to expression of exonucleases and superoxide dismutase, all with the purpose of minimizing oxidative damage (11, 201).
23 In the heat shock response, heat-shock proteins (HSPs) are expressed when bacteria are confronted with environmental stressors like temperature, high salt, and exposure to heavy metals (194). HSPs include molecular chaperones like DnaJ, DnaK, and GroEL, as well as ATP-dependent proteases like ClpP. Interestingly, stresses that induce expression of these HSPs in B. subtilis and E. coli, such as DNA damage, oxidative stress, and high osmolarity, do not induce them in S. pneumoniae (30, 223). In several pneumococcal strains, HSP ClpL, an ATPase, was shown to induce mRNA expression and production of cell wall synthesis protein PBP2x (222, 223). This led to decreased susceptibility to penicillin via thickening of the cell wall, as directed by PBP2x
(222, 223).
S. pneumoniae experiences many stressful conditions within the host that turn on its stress response pathways. Fluctuations in temperature (upper respiratory tract versus internal sites), pH and nutrient availability, as well as reactive oxygen species (ROS) produced by host phagocytes and potential antibiotic exposure all create a changeable, stressful environment within the pneumococcal niche (124). Several proteins have been demonstrated to be important in surviving particular stresses. The core pneumococcal genome codes for four Clp ATPases (ClpC, ClpE, ClpL, ClpX), which function in a strain-dependent manner as chaperones and regulators in response to stresses such as heat and competence induction (89, 124). ClpP-mediated proteolysis has a role in pneumococcal virulence, stress responses, and competence development (118). SpxA- like proteins are also believed to be involved in the development of competence and oxidative stress resistance (124, 226). Another regulatory system involved in responding to environmental stress is the two-component system (TCS) StpK-PhpP (163, 198). S.
24 pneumoniae stpK mutants are more sensitive to acidic, osmotic, and oxidative stresses because StpK is a positive regulator of genes involved in cell wall metabolism, DNA repair, and the oxidative stress response (124, 198). Other TCSs, ComDE and CiaRH, have been implicated in the development of competence (29, 40). Finally, surface- associated serine protease HtrA is involved in resistance to temperature and oxidative stress (88). It is also involved in genetic transformation, a proposed general stress response mechanism in S. pneumoniae (48, 181).
1.4.1 Oxidative stress response
Free radicals and ROS are highly reactive molecules whose levels increase as a result of environmental stresses such as desiccation. ROS are normal, unavoidable byproducts of cellular respiration that typically can be scavenged by dedicated enzymes
(e.g. superoxide dismutase) or neutralized by antioxidant molecules (14, 209). Oxidative stress often results from an imbalance of the endogenous ROS produced by a bacterial cell and the cell’s ability to either detoxify the reactive molecules or repair the damage they cause (54). Under desiccating conditions, oxidative metabolism and respiration may continue, particularly if water is still present within the bacteria or trapped against it by exopolysaccharides. ROS accumulate as a result of these ongoing cellular processes, and this build-up of toxic molecules can damage DNA, lipids, and proteins (16, 84). In particular, the Fenton reaction produces hydroxyl radical, which is highly reactive and therefore can attack and damage virtually all cellular components (124, 209). As a member of the upper respiratory tract microbiota, S. pneumoniae is exposed to oxygen
(O2) and, as a result, exogenous oxidative stress. Transcriptional repressor gene rgg is up-regulated under conditions of aerobic growth, and its inactivation leads to
25 pneumococcal sensitivity to O2 and paraquat (produces superoxide radical), but not H2O2
(20). S. pneumoniae also expresses two superoxide dismutases (manganese- and iron- dependent), which catalyze the conversion of superoxide to H2O2 and O2. Mutants in the manganese-dependent enzyme SodA display reduced growth under aerobic conditions and attenuated virulence in mice (253). Manganese is important in the oxidative stress response of the pneumococcus and other bacteria. Strains lacking PsaA (manganese transporter) and PsaD (thiol peroxidase) are sensitive to oxidative stress and have reduced expression of SodA (225). SpxB, the pneumococcal pyruvate oxidase, synthesizes acetyl-phosphate from pyruvate, phosphate and O2, a reaction that also results in carbon dioxide (CO2) and high levels of H2O2 (210) (Fig. 1-2). This acetyl-phosphate can be used as a source of ATP. S. pneumoniae has been measured to produce H2O2 at levels up to 1-2 mM in the culture supernatant (173). Intriguingly, although SpxB is responsible for nearly all of the H2O2 production by the pneumococcus, its presence is also critical for the bacteria to survive in the presence of H2O2 (174, 183). Mutants in spxB synthesize
85% less acetyl-phosphate than WT bacteria and their ATP levels decrease faster than
WT, so it has been suggested that the increased sensitivity of spxB mutants is due to low levels of ATP (174).
26
:-+6=*-%1#'@-?%A?,-(9-<;+,-0%
9-,+7% :1!% 1#;<+=*-%!,(-00% &'()*+,-% 3 1 >;";<%"-(/#;<+=/?% !"#$% 2 2% % &./0".+,-% 41 >;";<%.'<(/#'7+=/?% 2% !/<5 &(/,-;?%/#;<+=/?% 12% 56-,'78&%
Figure 1-2: Pneumococcal production of reactive oxygen species. Adapted from reference (84).
27 1.4.2 DNA damage repair
Another acknowledged aspect of desiccation tolerance, in addition to the general stress response, is an ability to overcome DNA mutations, as nucleic acids are targets for desiccation- and rehydration-induced damage (54, 179). Shibasaki’s group used E. coli to establish that drying causes DNA single- and double-strand breaks, and that DNA repair mutants ∆recA, ∆recB, and ∆uvrA∆recA are more sensitive to this desiccation than
WT bacteria (9, 10). Work on other bacteria has supported the findings from E. coli.
Exposing B. subtilis to drying via air or vacuum led to DNA lesions through strand breaks and DNA-protein cross-links, leading Dose and colleagues to conclude that efficient DNA repair is a key strategy for surviving dry conditions (43). Transposon mutagenesis of the nitrogen-fixing Gram-negative bacterium S. meliloti has shown that regulatory genes and uvrC, a DNA repair gene, are important in desiccation resistance
(87).
DNA damage induces the SOS response, wherein the cell cycle is stopped and
DNA repair occurs. During normal growth, the LexA repressor is bound to the operator of genes activated by the SOS response. When DNA damage occurs, RecA binds to the damaged single-stranded DNA (ssDNA), becomes activated, and interacts in turn with
LexA, inducing LexA autoproteolysis. Lacking LexA repression, the SOS genes are expressed. The uvr genes have weak LexA binding sites (SOS boxes) and so are expressed even upon low levels of DNA damage, thus causing nucleotide excision repair
(NER) to be the first SOS repair mechanism induced (145). NER works by excising damaged nucleotides from double-stranded DNA (dsDNA). As more DNA damage occurs, recA and other homologous recombination genes are expressed (93, 145).
28 RecBCD, with its helicase and nuclease activity, can unwind and nick dsDNA before loading RecA onto the 3’ single-stranded DNA (208). RecA initiates exchange between its DNA strand and one of an intact DNA duplex, and then this joint DNA molecule is resolved by RuvABC or dissociated by RecG (5). The RecF, or the RecFOR pathway, is an alternate means of homologous recombination, which is used to load RecA onto DNA in the event of ssDNA breaks or if RecBCD has been inactivated (153). S. pneumoniae has orthologs of RecFOR (100) and, in place of the Gram-negative RecBCD system, the pneumococcus uses RexAB for homologous recombination and double-stranded break repair (72). Homologous recombination enables the repair of lesions on ssDNA regions at replication forks by rendering these sites double-stranded and therefore able to be acted upon by NER (145).
In global genomic repair, the NER UvrABC excinuclease complex recognizes, binds to, and cleaves damaged DNA in an ATP-dependent process, thereby allowing
DNA ligase to replace the excised section. The UvrA dimer recognizes the site of damage and loads UvrB onto the DNA. After UvrA dissociates, UvrC binds the UvrB-
DNA complex and facilitates 3’ and 5’ incisions on the DNA (224). Both UvrC and this excised DNA are removed by DNA helicase II (UvrD). DNA polymerase I synthesizes new DNA to repair the gap, triggering the release of UvrB, and DNA ligase finishes the process by ligating the ends to the parental DNA. Mfd is important in transcription- coupled NER, where it identifies damage via recognizing a stalled RNA polymerase, and then binds the DNA, displaces the RNA polymerase, and loads UvrA onto the site of damage (224), after which the pathway continues as in global repair.
29 As a last resort, if NER and homologous recombination fail to fully repair the
DNA damage, the SOS response induces expression of umuC and umuD, which encode subunits of the mutagenic DNA repair polymerase V (Pol V) (145). Its main function seems to be the bypass of DNA lesions that block replication by the other polymerases, thereby allowing replication to continue at the cost of added mutations in the genome at these sites of DNA damage (62).
Two other DNA repair pathways are base excision repair (BER) and mismatch repair (MMR). BER occurs throughout the cell cycle and is responsible for removing small lesions from the chromosome, whereas NER removes large lesions that distort the dsDNA. Unlike the multi-protein NER complex, DNA glycosylases recognize and remove individual damaged or incorrect bases, leaving an abasic site that is cleaved by
AP endonuclease (129). This single-stranded break is repaired by polymerase and the new sequence is sealed by ligase. DNA MMR is used to repair nucleotide mismatches resulting from the two DNA strands slipping, leading to an incorrect base pairing and/or small loops of extrahelical nucleotides during replication (97, 116). Because there is no actual damage, just improper alignment, the MMR substrate differs from the damaged- nucleotide substrates of NER and BER. MutS, MutL (HexA and HexB, respectively, in
S. pneumoniae), and MutH are the E. coli MMR proteins. A MutS dimer recognizes and binds the mismatched base on a newly synthesized strand of DNA, and then a MutL dimer binds the MutS-DNA complex and activates MutH’s endonuclease activity (97,
116). MutH cleaves the DNA near the site of mismatch, and this nick allows MutL to load the DNA helicase and ssDNA-binding protein onto the DNA (116). These proteins
30 unwind and stabilize the ssDNA for digestion by exonucleases so that DNA polymerase
III can resynthesize the strand, which is sealed by DNA ligase (116).
1.4.3 Competence
Interestingly, S. pneumoniae lacks an SOS-like system (4, 31, 57), but instead has been shown to up-regulate its natural competence in response to stresses (181).
Competence and transformation are terms sometimes used interchangeably, however they are not actually synonymous. Bacterial transformation is a form of genetic recombination that occurs as a result of uptake and incorporation of exogenous DNA, whereas competence is the state of regulating transformation and other processes (48,
181). Of the more than 100 pneumococcal genes regulated by competence, less than 20% are required for transformation (48, 68, 176). Transformation (i.e. DNA uptake) was shown to be necessary for pneumococcal survival after DNA damage, however only the activation of competence, and not DNA uptake, was needed for S. pneumoniae to survive non-DNA-damaging stress conditions (48).
Natural competence in S. pneumoniae begins with environmental cues triggering comC to produce precursor competence-stimulating peptide (CSP), which is secreted by a dedicated transporter (98). CSP is processed during export through the ComAB transporter before this extracellular pool of CSP binds the ComDE TCS, the product of neighboring genes in the comCDE operon (98). Global regulators including the CiaRH
TCS control basal transcription from the comCDE operon (98). It is unknown what signal regulates pneumococcal ciaH activity, but it has been suggested that CiaRH is involved in the pneumococcal reaction to cell wall damage (40, 98, 142). The induction of competence leads to increased ciaRH transcription, and the CiaRH TCS has been
31 linked to control of early competence regulation, as well as normal cellular exit from the competent state (40, 98, 142).
The state of competence allows for transformation, and the late genes regulated by competence include those encoding proteins for the DNA uptake machinery as well as those triggering fratricide (98, 181). Competent S. pneumoniae cells have the ability to kill non-competent isogenic bacteria, an effect elicited by murein hydrolase CbpD (54,
98, 214). CbpD is expressed from a late gene promoter only during competence and, in conjunction with lytic enzymes LytA and LytC, it acts upon non-competent cells to trigger lysis and release of DNA (98, 104). While the precise details of the establishment of pneumococcal competence remain unknown, it has been demonstrated that competence can be induced as a response to stressful conditions, leading to lysis of neighboring cells, and thus providing immediate access to transforming DNA, which could be used to repair DNA damage resulting from the causative stress (48).
Transformation has been shown to occur between co-colonizing S. pneumoniae strains
(one with and one without a virulence factor), enhancing nasopharyngeal survival of the strain lacking the virulence factor (137).
1.4.4 Protein protection
Desiccation tolerance has long been attributed primarily to the ability to fix DNA damage and consequently to the function of DNA repair proteins. Recently, an analysis of the sequence, structure, and function of varied DNA repair proteins found no relationship between the protein sequence and the bacterium’s ability to withstand desiccation or radiation (41, 52, 167). Instead, a correlation was found between desiccation resistance and protection of proteins from oxidation (52). ROS produced
32 during dehydration cause modification to proteins in the form of carbonyl groups, but a high intracellular manganese:iron concentration can reduce this damage (52, 167).
Manganese ions (Mn2+) scavenge superoxide, and intracellular Mn2+ accumulation has been demonstrated to increase oxidative stress tolerance (192). It is believed that protein complexes containing Mn2+ scavenge radiation- and desiccation-induced ROS, thereby reducing the cellular concentration of iron-dependent radicals (52, 167) that could be otherwise used in the Fenton reaction.
It is more than likely that the ability to protect against and adequately respond to both protein oxidation and DNA damage is essential to bacterial desiccation tolerance.
These processes, in fact, do overlap to an extent. Due to the sensitivity of strains lacking psaA, PsaA, the binding component of a high-affinity Mn2+ transporter, is believed to be involved in oxidative stress resistance (101, 166, 225). Additionally, transcriptome analysis has implicated PsaA in pneumococcal competence induction (166), which, as mentioned above, can be used to alleviate the effects of DNA damage (48).
1.4.5 Osmotic stress response
There are two general phases in the bacterial osmotic stress response. First, there is a dramatic increase in potassium ion (K+) transport from the extracellular medium via turgor-responsive transport systems, as described in E. coli, Salmonella spp., and B. subtilis (130, 144, 242). This is followed by an increase in the synthesis of molecules such as glutamate or proline, which act as counter-ions for the increase in positive charges in the cytoplasm resulting from K+ import (130, 242, 251). Second, soluble zwitterionic molecules such as glycine betaine, proline, and carnitine, accumulate as a result of de novo synthesis or active transport (107, 130, 242). Under conditions of high
33 salinity, osmoprotectants increase the cytoplasmic volume and cellular free water content, enabling the cells to maintain a homeostasis under osmotic stress (130, 182). Like other stress responses, the response to osmotic stress is multivariate, as demonstrated by the induction of 125 B. subtilis genes upon bacterial transfer to a high salinity medium (213).
One of the ways bacteria respond to changes in environmental conditions is by adapting their membrane lipid composition (8, 175). Within the bilayer, lipids can change positions, allowing for membrane fluidity. Changes in lipid composition, including the ratio of saturated to unsaturated fatty acyl chains, chain length, and the degree of branching (8), can also affect membrane fluidity. Unsaturated fatty acids have a kink in their hydrocarbon tail chain, while saturated fatty acids have straight tails (42).
Therefore, the presence of more saturated fatty acids enables greater compression of the membrane because these fatty acids can crowd tightly together within the bilayer.
Pneumococcal fatty acid composition varies based on serotype and capsular phase variation (8).
Changes in membrane composition occur when bacteria are faced with heat stress, oxidative stress, and osmotic stress. Cardiolipin levels in the E. coli and B. subtilis plasma membrane go up upon entry into stationary phase as a result of increasing transcription of the gene encoding cardiolipin synthase (designated cls in the these bacteria) (190). Transcription of cls is elevated two- to three-fold when cells are under osmotic stress (79, 130, 190), and B. subtilis ∆cls strains show impaired growth in high- salt medium (191), while E. coli up-regulates cls transcription when in low osmolar medium (79). Cells lacking cls show higher proportions of PG, which further increases upon osmotic stress (130, 190, 191). Osmotic stress has also been shown induce
34 expression of B. subtilis pgsA, encoding PG synthase, leading researchers to suggest that osmotic regulation controls the relative proportions of zwitterionic (PE) and anionic (PG, cardiolipin) membrane phospholipids (130, 190).
It is likely that the dramatic changes in lipid composition during stress modify the membrane structure, affecting the regulation and activity of enzymes and transporters
(130). By interacting with the changing membrane lipids, membrane-bound mechanosensitive channels indirectly sense and respond to changes in the osmolarity of the cellular environment that cause membrane stretching (122, 190). If bacteria undergo an osmotic downshock, i.e. switching from a high to a low osmolar solution, the cells will swell to take in water and will also try to expel solutes, all in an attempt to reduce turgor pressure (115). First, the MscS (mechanosensitive channel of small conductance) heptamer opens to facilitate solute efflux. Then, under continued osmotic pressure, the
MscL (mechanosensitive channel of large conductance) pentamer opens for rapid discharge of cellular components (115, 140, 245). The E. coli MscL pore diameter expands 15-16 Å to a fully open diameter of 30 Å (38), while MscS is approximately 16
Å across when open (216). MscL has no ion preference, however MscS displays a weak preference for anions over cations (38, 46, 140). Individual E. coli mutants lacking either mscS or mscL are still able to respond to osmotic stress, but double mutants lyse upon hypotonic shock (85, 125). In addition to MscS and MscL, the E. coli genome encodes mini channel MscM and K+-dependent channel MscK (115). MscM has the lowest threshold for osmotic downshock, and so opens first, followed by MscK and MscS, with half the threshold of last-to-open MscL (115). This allows the bacteria to provide a tiered response to varying osmotic challenges and turgor pressure changes. B. subtilis possesses
35 one copy of mscL and three copies of putative mscS homologues (85). A quadruple mutant has no growth defect in high osmolar medium, while a ∆mscL strain is killed more rapidly than WT upon osmotic downshock (85).
1.5 Pneumococcal desiccation tolerance
Research dating to the 1890’s has investigated the ability of S. pneumoniae to withstand desiccation, however various desiccation techniques, surfaces, environmental conditions, and co-desiccated materials do not allow for easy comparisons between studies. In 1984, Mitscherlich and Marth compiled a comprehensive list of previous environmental survival experiments (149). Depending on the experimental conditions, the pneumococcus was shown to survive for anywhere from less than one day to 455 days. When exclusively considering experiments in which the bacteria underwent desiccation, this range shortens – less than one day to 140 days. Survival depends on temperature (bacteria at 4.4˚C survive longer than those at 26.7˚C), photo-oxidative exposure (bacteria in the dark survive longer than those in daylight), moisture (bacteria kept in dry conditions survive longer than those in moist conditions), and a protective matrix (bacteria in dried human sputum survive longer than those in dried rabbit blood)
(149). In the 30 years since this survey, little work has continued this line of experiments, yet an increased awareness of desiccation-tolerant pathogens, many of which are causative agents of nosocomial infections, necessitates further evaluation of S. pneumoniae’s potential to survive on and be transmitted by fomites.
To fill the gaps in our understanding of pneumococcal desiccation tolerance, it is necessary to more thoroughly evaluate this phenotype and identify mechanisms involved
36 in environmental persistence. With this knowledge, we can better evaluate risk factors for a fomite model of pneumococcal transmission and determine steps needed to reduce this possibility.
37 CHAPTER 2: Characterization of the Desiccation Tolerance Phenotype of Streptococcus pneumoniae
Portions of this chapter were previously published in: Walsh, R. L. and Camilli, A. 2011. Streptococcus pneumoniae is desiccation tolerant and infectious upon rehydration. mBio 2(3): e00092-11.
38 2.1 Overview
Commonplace activities such as talking, coughing, and sneezing disseminate large amounts of bacteria into the environment (74, 250), and research on respiratory tract pathogens has implicated dust and disintegrating sputum as reservoirs of bacterial transmission (236). Fomites, or environmental surfaces, have been described as probable or confirmed sources of infections caused by S. aureus, C. trachomatis, Klebsiella pneumoniae, A. baumanii, P. aeruginosa, various enterococci, and other bacteria (1, 96,
148, 158, 162, 202). Recent comprehensive literature searches on microbial persistence on dry surfaces identified more than 30 types of clinically relevant bacteria, including S. pneumoniae, that survive for anywhere from 30 minutes to over 30 months (113, 236).
Environmental survival often hinges on an organism’s ability to withstand periods of desiccation. “Animalcule” recovery from desiccation was first described by Antonie van Leeuwenhoek over three centuries ago (37, 106, 180). At present, the best-known and most-studied microbes that tolerate extended periods of desiccation are the cyanobacteria, extremophiles, and spore-formers (16, 17, 179, 180, 204). If S. pneumoniae were shown to survive desiccation, this bacterial durability could change our understanding of its transmission. As the only known reservoir of S. pneumoniae is the human upper respiratory tract, few studies have examined its capacity to persist in the environment. Confounding issues in interpreting these prior data including the following:
1) different dissemination strategies were used between studies, and 2) much work involved the direct desiccation of patient samples rather than bacteria alone (149).
Here we characterize S. pneumoniae’s capacity to survive desiccation and explore the potential for fomites as a transmission source. We developed a desiccation protocol
39 and used it to study S. pneumoniae viability over periods of desiccation ranging from 1 hour to 28 days. We found that desiccation and starvation are separate processes and desiccation tolerance likely is a species-wide phenomenon of the pneumococcus that does not depend on the presence of the polysaccharide capsule. Not only can S. pneumoniae survive extended periods of desiccation under ambient conditions, but also it retains its ability colonize, as assessed by the murine nasopharyngeal model.
40 2.2 Results
2.2.1 Design of a novel desiccation assay
To examine the ability of S. pneumoniae to withstand dehydration, we developed a desiccation protocol (Fig. 2-1). Preliminary work in our lab had utilized PCR tubes as the experimental fomite for a desiccation assay. S. pneumoniae was plated for overnight growth on blood agar and the bacterial lawn was transferred to a PCR tube for desiccation, but we were concerned that this might lead to uneven spreading, and thus drying, of the bacteria on the tube’s surface. As a result, we began by testing different fomites to determine which surface and spreading technique led to consistent bacterial drying and recovery. After testing PCR tubes, 1.5 mL Eppendorf tubes, 15 mL conical tubes, and petri dish lids, we determined that bacteria spread thinly onto petri dish lids using a plastic straight-edge dried quickly and showed higher levels of post-desiccation viability as compared with those from the other fomites. Additionally, we found that bacteria scraped off of the overnight blood agar plate survived desiccation at higher levels than those washed off the plate.
In our standard protocol, a wild-type (WT) encapsulated S. pneumoniae strain
D39 (serotype 2) starter culture was grown overnight on blood agar, scraped off of the plate surface en masse using the straight-edge, and evenly divided into separate pools, which were spread thinly onto polystyrene petri dish lids. The bacteria from one lid were immediately resuspended in Todd-Hewitt broth supplemented with yeast extract (THY) for initial time point (T0), and the remaining lids were desiccated in the dark under ambient conditions for predetermined times prior to rehydration with 1.5 mL THY and use of a tissue culture cell scraper to resuspend the bacteria. The number of viable cells
41 per lid was assayed by plating serial dilutions on blood agar and counting the resultant colonies, and the percent survival was calculated by dividing the time point by the T0 viable count.
42
I.,.2%& ;77%2.#$%"9& (%'*'4%-2&.-& I."*$%&#-2& <=>&1(+$0&6+(& 4"#$%&6+(&AJK& E%'*'4%-2& #F%(&G&2#9'&6+(& A"+'%&+,%(& E%'*'4%-2& 2.'0&1+B+7C& #F%(&:&2#9'&6+(& 4"#)%&.-&2#(D& <:2&@7%4+.-$& 3)(#4%&1#)$%(.#"&"#5-& #$&(++7&$%74& 6(+7&4"#$%8'&'*(6#)%& E%'*'4%-2& #F%(&H&2#9'&6+(& < &@7%4+.-$& 34".$&%,%-"9&.-$+&:&4++"'&#-2& H2 '4(%#2&%#)0&#'&#&7+-+"#9%(& +-&4+"9'$9(%-%&4%$(.&2.'0&".2& Figure 2-1: Schematic of desiccation assay. S. pneumoniae starter cultures were thawed and plated for overnight growth on blood agar. After growth, the bacterial lawn was scraped off the plate, evenly divided and spread thinly onto polystyrene petri dish lids. The bacteria from one lid were immediately resuspended in THY broth for initial time point T0. The remaining lids were desiccated in the dark under ambient laboratory conditions for times ranging from 1 hour to 28 days. The image illustrates three potential time points. Upon resuspension, the number of viable cells per lid was assayed by plating serial dilutions on blood agar and counting the resultant colonies. The percent survival was calculated by dividing the time point viable count by the T0 viable count. 43 2.2.2 S. pneumoniae can survive periods of desiccation of at least 28 days Using this standard desiccation assay, we desiccated S. pneumoniae for increasing amounts of time, ranging from 1 hour to 28 days. After one day of desiccation, 10% of the cells remained viable (Fig. 2-2A). After one week of desiccation, 0.1-1% of the bacteria remained viable. We attribute the spread in survival within a single time point to inherent variability in the assay – in splitting and spreading the bacteria on the fomite, as well as in day-to-day changes in environmental conditions. Overall, our results show that plate-grown S. pneumoniae survives desiccation periods of at least four weeks at ambient temperature and humidity. Similar to other bacteria (45, 58, 82, 94, 151, 227), the pneumococcus demonstrates two rates of decay in its survival (Fig. 2-2B) – most of the population dies within four days, and a small subset are able to persist for weeks longer with a slower rate of loss. 44 !" 100 10 1 l a 0.1 v i v r u S t 0.01 n e c r e P 0.001 0.0001 0.00001 0.000001 1 day 1 hour 2 hours 4 hours 6 hours 8 hours 2 days 3 days 4 days 7 days 10 hours20 hours 14 days 28 days Time of Desiccation #" 100 10 1 l a v i v 0.1 r u S t 0.01 n e c r 0.001 e P 0.0001 0.00001 0.000001 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Time (Days) Figure 2-2: S. pneumoniae D39 survival after desiccation. Bacteria were resuspended and plated after 1 hour to 28 days of desiccation to determine viability. Data were pooled from multiple biological replicate experiments. A, All data plotted, where open circles represent individual samples, and bars show the medians. B, Medians plotted on a time- scale with lines drawn to represent two survival decay rates. 45 2.2.3 Different growth states support subsequent desiccation survival In the desiccation assay described in Section 2.2.2 using colonies scraped from blood agar plates, the starting population was likely to be very heterogeneous with respect to growth rate and, thus, physiological state. To examine a more homogenous starting population, we tested exponentially growing S. pneumoniae cells from broth culture for desiccation over a seven-day period (Fig 2-3). At the one- and two-day time points, the cells desiccated after exponential phase growth survived at significantly higher levels than those desiccated after overnight plate growth (Fig. 2-4). However, the advantage in survival was no longer observed by four and seven days (Fig. 2-4), suggesting that the physiological growth state does not necessarily dictate an ability to survive longer periods of desiccation. 46 !""#$%&$'($)*+,*-"$ ./,0$1,2-,"-$)*+,*-"$$ 9+/:*4,$";"7+<$/7,4$,*="1$ ?"1*16"7#$/7$@&$A($BCD$$ 3-4.$,4$ !6/7>$-"'4;"$ "5647"782+$6021"$ 1*6"-72,27,$ ?"1*16"7#$E4-$B&$ G-<$H&$'/7$ /7$044#$ B-271E"-$F$16-"2#$ 47,4$6",-/$+/#1$ G"1/))2,">$ -"1*16"7#$2,$ %>$I>$J$F$K$#2<1$ Figure 2-3: Schematic of exponential growth-phase desiccation assay. A starter culture of S. pneumoniae was thawed and used to seed a 10 mL culture of THY + Oxyrase. This was grown to mid-exponential phase, as measured by OD600 = 0.4, and then aliquoted into 500 µL volumes. The aliquots were spun, resuspended in 50 µL THY, and this slurry was transferred and spread onto a polystyrene petri dish lid. The lids were placed in the biosafety cabinet with air-flow on for 30 minutes to quickly dry excess liquid, at which point the T0 sample was resuspended in THY for dilution and plating. The remaining lids were desiccated in the dark under ambient laboratory conditions for one, two, four, or seven days, at which point they were similarly resuspended, diluted, and plated for CFU. The percent survival was calculated by dividing the time point viable count by the T0 viable count. 47 ** ** 100 10 l a v i v r u S t 1 n e c r e P 0.1 0.01 1 2 4 7 Time of Desiccation (Days) Plate-Grown Exponential Figure 2-4: Plate-grown versus exponential growth-phase cell desiccation. After overnight plate growth (circles) or exponential phase growth (diamonds), S. pneumoniae was desiccated for 1, 2, 4, or 7 days under ambient laboratory conditions. ** p < 0.01 by Mann-Whitney U test 48 2.2.4 Desiccation and starvation are separable stresses Numerous forces that contribute to bacterial cell death over time are at work during desiccation (16, 127, 179). It is possible that cell death of S. pneumoniae is due primarily to nutrient deprivation, regardless of other factors. To test this, we conducted simultaneous experiments under two conditions: 1) desiccation, and 2) starvation but with maintenance of hydration. The desiccated samples were treated according to our standard desiccation protocol, while the starved samples were spread onto phosphate- buffered saline (PBS) agar plates rather than polystyrene petri dishes. All samples were placed in the dark under ambient conditions for 6, 24, or 48 hours prior to collection for comparison to the T0 viable count. We saw a significant difference in bacterial recovery between the desiccated and starved samples (Fig. 2-5), with the starved samples losing viability at a much higher rate than the desiccated samples. The starved samples underwent a shift from slightly higher survival after 6 hours (p < 0.05) to lower survival at 24 hours (p < 0.05) and 48 hours (p < 0.001), confirming that nutrient deprivation is a different phenomenon than dehydration. 49 *** * ** 100 10 l a v i v r u S t 1 n e c r e P 0.1 0.01 6 24 48 Time (Hours) Desiccated Starved Figure 2-5: S. pneumoniae D39 survival after desiccation versus nutrient deprivation. Bacteria were recovered 6, 24, or 48 hours after desiccation (closed circles) or starvation on PBS agar (open circles), and viability was determined. Circles represent biological replicates (outliers measured by Grubbs’ test were removed), and bars show the medians. The probability that medians differ at each time point is shown by asterisks. * p < 0.05, ** p < 0.005, *** p < 0.0005 by Mann-Whitney U test 50 2.2.5 S. pneumoniae desiccation tolerance is not dependent on the polysaccharide capsule As research has indicated that Gram-positive bacteria survive desiccation better than their Gram-negative counterparts (69, 94, 152), we wondered if the polysaccharide capsule of S. pneumoniae contributes to its ability to withstand desiccation. A capsule is present in essentially all S. pneumoniae clinical isolates and is required for efficient host colonization as well as invasive disease (2, 159, 239). To assess the role of capsule in desiccation tolerance, we compared the survival of D39 to that of an acapsular derivative, AC326. We saw no significant difference in bacterial viability at any time point at up to one week of desiccation (Fig. 2-6), leading us to conclude that the capsule is not an important factor in surviving desiccation. An anti-type 2 capsule Western blot demonstrated the presence of the capsule in cells desiccated for up to seven days, confirming that the lack of phenotypic difference between D39 and the acapsular strain is not due to D39 down-regulating production of the capsule, and thus appearing acapsular, during desiccation (Fig. 2-7). 51 100 10 l a v i v r u S t 1 n e c r e P 0.1 0.01 4 8 24 48 72 96 168 Time (Hours) of Desiccation + capsule - capsule Figure 2-6: Desiccation tolerance of S. pneumoniae D39 (encapsulated) and its acapsular derivative AC326. Bacteria were recovered 4 to 168 hours (one week) after desiccation. Data for D39 with the capsule (white) and without the capsule (gray) are shown as the median values (n = 4), and bars represent the ranges. 52 !"#$%&'()%)*+,- .$#+/-0%1"22+3%#- 4$#+/-0%1"22+3%#- 5$6/'%-.-7+'18,%- Figure 2-7: Anti-capsule Western blot of desiccated S. pneumoniae. Mid-exponential cells and two- and seven-day desiccated WT D39 S. pneumoniae were probed with an anti-type 2 capsule antibody. 53 2.2.6 Desiccation tolerance is a property shared by diverse pneumococcal strains S. pneumoniae has a highly plastic genome, with up to 10% of variation in gene content between strains (71, 78, 119, 229). Since we eliminated the capsule as the basis for desiccation tolerance, it is possible that some genetic specificity of the D39 strain enhances its ability to survive desiccation. To determine if desiccation tolerance is shared or not by strains of other serotypes, we performed 48-hour desiccation experiments on 17 strains representing 14 different serotypes. All 17 strains tested survived desiccation at viabilities ranging from 0.1 to 10% (Fig. 2-8). For three of the serotypes (6A, 6B, and 18C), we tested both opaque and transparent colony phase variants and saw no correlation between phase and desiccation tolerance. As capsular polysaccharide expression differs greatly between these two phase variants and is related to virulence in a mouse model (109), this supports our previous conclusion that the capsule is not a key factor in pneumococcal desiccation tolerance. Additionally, our experiments indicate that the ability to withstand the stresses of desiccation is a trait shared by numerous strains of S. pneumoniae. 54 100 10 l a v i v r u S t 1 n e c r e P 0.1 0.01 1 2 3 4 5 8 14 15 9V 19F 23F 6A opaque 6B opaque 18C opaque 6A transparent 6B transparent 18C transparent Serotype Figure 2-8: Desiccation tolerance of 17 S. pneumoniae strains. Bacteria were recovered after 48 hours of desiccation. Each data point represents an independent experiment, and bars indicate the medians. 55 2.2.7 Desiccated S. pneumoniae retain the ability to colonize after rehydration To extrapolate the importance of fomites as a source of S. pneumoniae transmission, it first must be established that, upon rehydration, desiccated bacteria are capable of colonization. Because S. pneumoniae is naturally virulent in mice and can colonize the nasopharynx asymptomatically, as in humans, we used this murine model to test whether desiccation affects colonization. We intranasally inoculated 8- to 12-week- old female Swiss-Webster mice with 5 µL/nare of PBS (mock), PBS-suspended non- desiccated S. pneumoniae D39 (grown overnight on blood agar plates) or PBS-suspended 48-hour-desiccated S. pneumoniae D39 (Fig. 2-9). Three days post-inoculation, the mice were killed, and the nasal lavage fluid was plated on blood agar containing 3 µg/mL Gentamicin (Gent), to which S. pneumoniae is naturally resistant. The mock group had no detectable S. pneumoniae colonies, but the bacteria desiccated for 48 hours colonized the murine nasopharynx well (Fig. 2-10). In the non-desiccated WT group, 80% of mice (4/5) had detectable levels of colonization, while 75% of mice (6/8) in the desiccated group were measurably colonized. The relatively low dose of inoculum used (~1.5 x 104 to 3.2 x 104 CFU/mouse) likely increased the spread seen in the data, as the mice had a better chance of clearing the bacteria than if the dose used was higher. Although the median load of bacteria recovered was higher for the non-desiccated challenge group, there was no significant difference in colonization levels between these mice and the desiccated group (p = 0.27), indicating that desiccation does not exert a major negative impact on the ability of S. pneumoniae to colonize hosts. 56 R*+;( F*"7D2&)++%#2<( D2&)++%#2<( @'%#2(&#%$#2$(+,'#,$2( *A2$")BC#(*"(?'**<(%B%$( D2&)++%#2( E6(C*,$&( =+$%G2(,GN( &G$2%<(*"( G2#$)(<)&C(')<( 8H(@I=( >%&CN(&,&G2"<()"(8H(@I=( =,&G2"<()"(8H(@I=( !"#$%"%&%'()"*+,'%-*"(./(0'1"%$23(*4(425%'2(67897:22;7*'<(=:)&&7>2?$(5)+2( J(<%K&( L,#C%")M2N("%&%'('%A%B2N(G'%#2(4*$(OPQ( Figure 2-9: Schematic of murine nasopharyngeal colonization protocol. S. pneumoniae D39 starter cultures were plated for overnight growth on blood agar. After growth, the bacterial lawn was either washed and resuspended in PBS for inoculation (non-desiccated) or was scraped off the plate and desiccated for 48 hours prior to resuspension in PBS (desiccated). Female 8- to 12-week-old Swiss-Webster mice were intranasally inoculated with 5 µL/nare of either PBS (mock), PBS-suspended non- desiccated S. pneumoniae (non-desiccated), or PBS-suspended 48-hour desiccated S. pneumoniae (desiccated). Three days after inoculation, the mice were euthanized, nasal lavage was performed, and this was plated onto blood agar plates containing 3 µg/mL Gent for colony enumeration. 57 p = 0.27 105 104 103 L m / U F C 102 1 10 Limit of Detection 100 Non-Desiccated Desiccated Mock Mouse Group Figure 2-10: Murine nasopharyngeal colonization by desiccated versus non- desiccated S. pneumoniae. Shown are bacterial loads in nasal lavage fluid at three days post-intranasal inoculation with 1.5 x 104 CFU (non-desiccated) and 3.2 x 104 CFU (desiccated) S. pneumoniae D39. Circles represent individual mice, and bars indicate the medians (p value determined by Mann-Whitney U test). 58 2.3 Discussion To our knowledge, this is the first study to thoroughly characterize the desiccation tolerance of S. pneumoniae under ambient conditions. Using a polystyrene surface as a fomite model, we demonstrate the environmental survival of S. pneumoniae over a period of four weeks. Although there was a decline in bacterial survival over time, the median total concentration of viable cells, even after two weeks of desiccation, fell within a previously suggested 50% infective dose (ID50) range for this pathogen to cause IPD (200, 252). Since more bacteria are generally needed for invasive disease than for intranasal inoculation, this number may be sufficient for nasopharyngeal colonization, the first step toward infection (19). However, numerous variables will likely impact the frequency of transmission via fomites in natural settings. It is important to make a distinction between the process of desiccation tolerance and its intrinsic nutrient limitation. It could be argued that the death of S. pneumoniae over time on fomites is simply a result of starvation and not due to stresses associated with desiccation and rehydration. Nonetheless, two points argue against this hypothesis. First, we saw significantly less longer-term recovery of starved but hydrated cells versus that of desiccated cells, which indicates that nutrient deprivation and desiccation are distinct stresses. Our data are further supported by results from a previous study, in which S. pneumoniae mixed with dust, pus, and blood was viable approximately twice as long under dry versus moist conditions (149). Second, despite the absence of nutrients on fomites, we were able to recover viable cells after long time periods (up to 28 days) of desiccation. This suggests that the bacteria enter into some kind of stasis, enabling them to tolerate the stresses of desiccation. Desiccation may trigger an evolved survival 59 pathway of S. pneumoniae, or survival may simply be attributed to a fortuitously protective cellular architecture. An investigation of the bacterial factors that mediate desiccation tolerance should be able to distinguish between these possibilities. It is possible that bacteria in stationary phase are at an advantage when faced with desiccation because they are already in a metabolic stasis, unlike compared to exponentially growing cells. Our results show that both mid-exponential and plate-grown S. pneumoniae are able to survive desiccation at comparable levels after four and seven days, which suggests that growth state may not affect long-term desiccation tolerance. We are not sure, however, of the growth state of the plate-grown bacteria, and postulate that this group is comprised of bacteria in various metabolic stages. This would explain both the large spread seen in the recovery of plate-grown bacteria, as well as why our data do not cluster with any desiccated group (exponential planktonic, stationary planktonic, or biofilm bacteria) tested by Marks et al (138). Using two strains of S. pneumoniae, including D39, and two strains of S. pyogenes, they found that biofilm streptococci survive desiccation at high levels and for longer periods of time than either mid-exponential or stationary planktonic cells. Perhaps other characteristics of biofilms, such as the protective extracellular matrix, contribute to bacterial survival during desiccation. Numerous species of Gram-negative and Gram-positive bacteria have been shown to persist under desiccating conditions, although the literature suggests that Gram- positive bacteria exhibit enhanced tolerance to dry conditions (94, 152). One factor shared by many Gram-negative and Gram-positive bacteria is the presence of an extracellular polysaccharide. In S. pneumoniae, the polysaccharide capsule is a virulence 60 factor, helping cells penetrate the mucus layer overlaying mucosal epithelia and escape phagocytosis (19, 109, 159). Using an acapsular mutant strain, we demonstrated that the capsule is not required for S. pneumoniae D39 desiccation tolerance. This corresponds with the results from our desiccation experiment on serotypes 6A, 6B, and 18C, in which we observed no clear trend associated with capsular phase variance. The variants of serotype 6A demonstrated near-identical recovery, while the other two serotypes displayed opposite recovery patterns between opaque and transparent variants. Similar variation in desiccation tolerance was seen across other S. pneumoniae serotypes, including those seven previously identified as the most common in cases of IPD globally (99). We observed recovery across all 17 tested strains, accounting for 14 unique serotypes. Thus, a resistance to the stresses of desiccation appears to be a phenomenon intrinsic to the S. pneumoniae species. This is especially interesting, considering the wide genetic diversity among S. pneumoniae strains. Its natural transformable ability, combined with its high rate of recombination, enables S. pneumoniae to adapt readily to both antibiotic and vaccine selective pressures (36, 80, 103). Several years after the introduction and widespread administration of the heptavalent polysaccharide capsule vaccine (PCV7), IPD caused by the seven vaccine serotypes markedly decreased (39, 81). Simultaneous with this decrease was the emergence of multidrug-resistant non-vaccine serotypes, especially multidrug-resistant 19A, as leading causes of IPD (36, 80, 171, 177). Our data support the hypothesis that desiccation tolerance is a species-wide phenomenon; therefore this ability is vertically transmitted in S. pneumoniae and likely would not be gained or lost as a result of its frequent horizontal transfer events. 61 It has been noted that increased post-desiccation recovery rates are seen when bacterial cells are more concentrated at the time of dissemination (113, 141, 179, 237), and so it is conceivable that the bacterial viability in our experiments is artificially high based on the density of bacteria spread on the polystyrene surface (2.2 x 108 CFU over an area of ~600 mm2). In experiments testing the number of beta-hemolytic streptococci expelled by respiratory activities, Hamburger and Green showed that nose blowing, more than coughing or sneezing, forces out the most bacteria (74). In patients with beta- hemolytic streptococcus-positive nose cultures, nose blowing resulted in an average expulsion of 1.1 x 107 CFU and a maximum of over 109 CFU (74); therefore, it is possible that an S. pneumoniae (alpha-hemolytic streptococcus) carrier could expel concentrations approximating those used in our experiments. Additionally, they recovered an average of 7.9 x 105 CFU from the hands of nasal carriers three hours after they were last washed and found that these individuals transferred beta-hemolytic streptococci to other surfaces they touched (74). These surfaces, as well as dried handkerchiefs used during nose blowing, served as major sources of airborne beta- hemolytic streptococci (74). A recent study on experimental human pneumococcal carriage using serotypes 6B and 23F showed that nasal colonization could be established with 104 to 105 bacteria (64). It is therefore possible that, even with bacterial loss during desiccation, starting concentrations of as little as 107 pneumococci would allow for enough bacteria to survive several days to several weeks in the environment before subsequent transfer to and colonization of a new host. Fomites have been linked to the transmission of other Gram-positive, non- sporulating pathogens. MRSA was shown to survive 1 to 90 days on common hospital 62 materials (158) and has been proposed to cause infections via direct fomite-to-person transmission (149). Twenty weeks after desiccation of blood containing group A streptococci, growth in fresh blood indicated that the bacteria retained both viability and virulence (184). The hypothesis that S. pneumoniae may survive in the environment and use fomites as a source of transmission is not unprecedented. In fact, Walther and Ewald’s “sit and wait” hypothesis, which predicts that virulence correlates with durability in the external environment, identified a high-virulence, high-survival group of human respiratory tract pathogens that includes variola (smallpox) virus, Bordetella pertussis, Mycobacterium tuberculosis, C. diphtheriae, and S. pneumoniae (236). Here we tested this hypothesis by investigating the capacity of desiccated S. pneumoniae to colonize the murine nasopharynx. Colonization at levels not significantly different from those of non- desiccated S. pneumoniae argues in favor of the hypothesis that fomites could serve as an alternate source of pneumococcal spread. Additional work points to fomite transmission. S. pneumoniae has been shown to survive several hours on human hands (74, 138), easily facilitating transfer to fomites – and various surfaces allow pneumococcal survival on a time scale of hours to months (149). Perhaps most convincing is a study indicating fomite-driven infections occurred during a 1946 pneumococcal outbreak in a military academy (83). We do not doubt that airborne respiratory droplets, transmitted person-to- person, account for the majority of new pneumococcal colonization events. Nonetheless, we have demonstrated long-term pneumococcal viability during desiccation, after which these bacteria are still capable of colonizing the nasopharynx of susceptible hosts. We propose, then, that fomites serve as reservoirs for S. pneumoniae. 63 CHAPTER 3: Identification and Physiological Characterization of Genes and Pathways Involved in Streptococcus pneumoniae Desiccation Tolerance 64 3.1 Overview To identify the underlying mechanisms of S. pneumoniae desiccation tolerance, we took a genetic approach. We constructed and desiccated a nearly saturated transposon insertion library, after which we performed the Tn-seq method (230), a high throughput sequencing and quantitation of the transposon junctions. We selected 12 genes to study further and put their individual gene replacement mutants through desiccation assays to confirm the genes’ roles in this process. Like in other bacteria, stress responses seem to be critically important in S. pneumoniae desiccation tolerance. We subsequently conducted stress assays to study the functions of these genes in pneumococcal physiology and better understand their involvement in desiccation. Phenotypes were seen in mutants encoding the Mfd and UvrABC nucleotide excision repair complex, SpxB (pyruvate oxidase), SodA (superoxide dismutase), MscS (mechanosensitive channel of small conductance), and cardiolipin synthase. We found that S. pneumoniae’s resistance to desiccation is linked to its ability to withstand various stresses, including those imposed by osmotic changes and ROS, leading to DNA damage that requires repair. 65 3.2 Results 3.2.1 Construction of a transposon insertion library We combined S. pneumoniae genomic DNA (gDNA) with plasmid DNA harboring the Mariner mini-transposon derivative Magellan6 encoding Spectinomycin (Spec) resistance and a hyperactive purified transposase in vitro. Magellan6 randomly inserts at TA dinucleotide sites in the gDNA, which was subsequently transformed into naturally competent S. pneumoniae, producing a population of bacteria in which each has one random transposon insertion in its DNA. Because it is more difficult to transform encapsulated pneumococci, we took two approaches to library construction. First, we performed multiple in vitro transpositions and transformations into WT D39, as described above. Second, we made gDNA from pre-existing 10,000-mutant transposon insertion libraries made by Tim van Opijnen in either the acapsular D39 (serotype 2) or acapsular TIGR4 (serotype 4) background, which we also transformed into WT D39. As S. pneumoniae more readily incorporates gDNA than transposition products, this method produced a high number of transformants. Because we began with a limited number of transposon insertion locations – 10,000 sites instead of any TA site in the genome – we conservatively estimated that only 50% of our transformants from method 2 contained unique insertion sites. In all, we combined multiple small libraries (Table 3-1) into an aggregate library, which we later determined via Tn-seq contains approximately 28,000 independent transposon insertions. This is nearly saturated for S. pneumoniae D39’s 2200-gene chromosome. 66 Table 3-1: Transposon library complexities. Transposon insertion libraries were created by transforming WT D39 with in vitro transposition products (experiments F, G) or with gDNA from either pre-existing acapsular TIGR4 libraries made by Tim van Opijnen (TvO) (experiments D1, D2, E), pre-existing WT D39 libraries made by Tim van Opijnen (experiments A, B, C), or pre-existing acapsular D39 libraries made by Tim van Opijnen (experiments H1, H2, H3, H4, H5, H6) or by me (RW) (experiment H7). Different libraries are designated L1-L6. The ‘Transformants’ column is a total count of transformant colonies on the selective blood agar containing 200 µg/mL Spec. 67 3.2.2 Two- and seven-day desiccation of transposon insertion library Two starter cultures of the aggregate library were thawed and plated onto large blood agar plates (154 cm2) containing 200 µg/mL Spec. After overnight growth, the lawn was scraped off of the plate and evenly divided into three pools, which were spread onto large (165 cm2) petri dish lids. The bacteria from one lid were immediately resuspended in THY for T0, and a small volume was used to dilute and plate for CFU. The resuspended bacteria were then split in half – half was centrifuged, washed in PBS, re-centrifuged and the supernatant (SN) removed prior to freezing the pellet (total cells); the other half was plated onto Spec-containing blood agar for overnight growth, the colonies were collected with THY, centrifuged, washed in PBS, and the SN removed prior to freezing the pellet (live cells). This process was repeated at 2 (T2) and 7 (T7) days post desiccation (Fig. 3-1). We plated half of the resuspended bacteria so that we could make gDNA from outgrown colonies and look for transposon insertions exclusively from cells that survived the desiccation selection (live cells) instead of from total gDNA, which may include dead cells on the fomite. To see whether or not there is difference between total-cell gDNA and live-cell gDNA, we ran both sets of samples on a gel. As suspected, we saw no difference in gDNA quality, as visualized by a distinct band, between the total- and live-cell samples for T0 (Fig. 3-2, lane 2 compared to replicate lanes 3 and 4). However, the smeared band of the total-cell gDNA for both T2 and T7 show that the gDNA had undergone degradation, unlike the live-cell samples (Fig. 3-2, lanes 6 compared to replicate lanes 7 and 8, lane 10 compared to lane 11). In addition, overnight outgrowth led to more 68 colonies, and therefore more gDNA to use in sequencing preparation. Consequently, we moved forward using the live-cell gDNA for all time points. 69 !"#$%&"'()#)*& +$#)$%)&,-"$-)%& 3,)#4%&(#,$%)'#"&"#70& ./%)0'12$&.0&"#)1%& 8).9&4"#$%:+&+-)8#,%& 34%,5,.0$#'0'01& ("..6#)&4"#$%& 34"'$&%/%0"*&'0$.&;&4.."+&+4)%#6&%#,2&#+&#&9.0."#*%)&.0&"#)1%&4."*+$*)%0%&4%$)'&6'+2&"'6& =G& =E6& =F6& <%+-+4%06&'0&=>?& H'"-$%&4"#$%&8.)&IBJ& 34"'$&/."-9%& 34'0&6.70&2#"8& !"#$%&2#"8&./%)0'12$& =.$#"&,%""+& @#+2&(#,$%)'#"& "#70&.A&4"#$%& @#+2&'0&!O3P&)%+4'0& D-$1).7&$.&+%"%,$& &8.)&"'/%&,%""+& B)%%C%&4%""%$& @#+2&'0&!O3P&)%+4'0& K#L%&1HMN& B)%%C%&4%""%$& 70 Figure 3-1: Schematic of desiccation of transposon insertion library. A starter culture of the WT D39 aggregate transposon library was thawed and plated for overnight growth on a large blood agar containing 200 µg/mL Spec. After growth, the bacterial lawn was scraped off the plate and evenly divided into three pools, each of which was spread thinly onto the lid of a large polystyrene petri dish. The bacteria from one lid were immediately resuspended in THY broth for initial time point T0. A small volume was taken for diluting and plating for CFU, while the remaining volume was split in half. Half (total cells) was centrifuged, washed in PBS, re-centrifuged, and the supernatant (SN) was removed before freezing the pellet. The other half was plated onto a Spec- containing blood agar plate to allow for outgrowth of live cells. The next day the plate was washed with THY, and this was centrifuged, washed in PBS, re-centrifuged, and the SN was removed before freezing the pellet. The remaining two petri dish lids were desiccated in the dark under ambient laboratory conditions for 2 (T2) or 7 (T7) days, at which point they were resuspended and handled as described for the T0 time point. The frozen pellets were used to make gDNA. 71 )4$48(##))#"7(56(##)8(9#"6&4()*))#"7(56(##)8(9#"6&4()2))4$48(##))#"7(56(##)8(9#"6&4()*))#"7(56(##)8(9#"6&4()2))4$48(##))#"7(56(##) 3 + 3 + 3 + 3 2? 3 2? 3 2? 3 =? 3 =? :&;(<) *) 2) 1) 0) /) .) =) -) >) *+) **) !"#$%&'(') *+,+) -,+) .,+) /,+) 0,+) 1,+) 2,+) *,/) *,+) +,/) Figure 3-2: Gel of total-cell and live-cell gDNA from T0, T2, and T7 library samples. Genomic DNA was prepared from desiccated aggregate library samples and 300 ng per sample was loaded onto a 0.8% agarose gel. Lanes 2, 6, and 10 are the gDNA from total- cell samples of the T0, T2, and T7 time points, respectively. Lanes 3 and 4 (technical replicates), 7 and 8 (technical replicates), and 11 are the gDNA from the live-cell samples of the T0, T2, and T7 time points, respectively. 72 3.2.3 Two- and seven-day Tn-seq sample preparation and Illumina sequencing data analysis Live-cell gDNA from before (T0) and after (T2, T7) desiccation selection was digested with MmeI, ligated to adapters with Illumina-recognizable barcodes, and PCR- amplified before being gel purified. The unique 6-base pair (bp) barcodes within the adapters allowed us to mix the six samples (three time points each from two biological replicates), run them on the same Genome Analyzer II (Illumina) flow cell lane, and subsequently separate them after sequencing. From the sequence data, we were able to assign fitness values to all genes represented in the transposon library as described (230). The genes with fitnesses significantly different from WT in the 2-day desiccation are listed in Table 3-2. After analysis, we determined that the desiccation bottleneck for the T7 samples was too high (nearly 90%) for us to rely on these data. In other words, there was so much loss over the 7-day desiccation period that it was not possible to reliably determine which genes had fitness defects due to their importance for desiccation tolerance versus which genes fell out of the population due to stochastic loss. Therefore, we only considered the T2 data when selecting genes to study further. 73 Table 3-2: Genes from the 2-day Tn-seq desiccation. Listed are genes with fitnesses significantly (white background) or nearly significantly (gray background) different from the WT fitness of 1 after analysis of the 2-day aggregate library desiccation. Genes with nearly significant fitnesses are those with p values immediately below the significance cut-off, which likely would be considered significant given more transposon insertions in the gene. (p value determined by Student t test with Bonferroni correction) 74 3.2.4 Four-day desiccation of transposon insertion library and sample preparation for Illumina sequencing We wanted another time point to evaluate in addition to the 2-day desiccation, so we conducted a 4-day desiccation of our aggregate library hoping that the bottleneck would not be too high to confound the results. To further reduce this possibility, we desiccated 20 replicates and combined the outgrown (live-cell) samples so that we had a higher starting number of bacteria to use to prepare gDNA. Instead of splitting each bacterial lawn into multiple pools and spreading onto petri dish lids, we scraped up a lawn and used all of it for a single time point. This was possible, as we expected each lawn was identical in strain composition to all others because they came from the same set of starter cultures. The T0 samples were scraped off the overnight plate, transferred to THY, mixed, and pelleted for gDNA. The T4 samples were scraped off the overnight plate and spread onto large petri dish lids for 4 days of desiccation, after which they were resuspended with THY, and plated onto Spec-containing plates to select for live cells. The following day, these plates were washed, and the samples were combined, mixed, and pelleted for gDNA. To combine the replicates, we pooled four T0 samples into two groups (two per group) for sequencing, and combined 20 T4 samples into five groups (four per group) prior to gDNA preparation. Instead of using the MmeI protocol to prepare the samples, we used the newly- created Homopolymer tail-mediated ligation PCR (HTML-PCR) method (121). Briefly, live-cell gDNA from before (T0) and after (T4) desiccation selection was sheared and poly-dC tails were added to the 3’ ends using Terminal deoxynucleotidyl Transferase (TdT). Nested PCR was used to amplify out from one end of the transposon and a poly- 75 dG-containing reverse primer was used to amplify the shear-generated end of DNA template molecules. Only templates in which the poly-dC-modified sheared end was reasonably close to the correct end of the transposon were amplified in PCR. The primers added on Illumina-specific sequences required for sequencing on the Illumina platform, and also added on Illumina-recognizable 8-bp barcodes. The genes with fitnesses significantly different from WT in the 4-day desiccation are listed in Table 3-3. 76 77 Table 3-3: Genes from the 4-day Tn-seq desiccation. Listed are genes with fitnesses significantly (white background) or nearly significantly (gray background) different from the WT fitness of 1 after analysis of the 4-day aggregate library desiccation. (p value determined by Student t test with Bonferroni correction) 78 3.2.5 Selection of genes involved in desiccation Fitness determined by Tn-seq is a measure of a gene’s importance for surviving a selection process – in this case, desiccation. Genes uninvolved in the process have WT, or neutral, fitness values of 1, genes important for surviving desiccation have values below 1, and genes disadvantageous in desiccation have values above 1. The 2-day analysis resulted in 23 genes with fitnesses significantly different than 1 (Table 3-2), while the 4-day experiment resulted in 55 genes (Table 3-3). In selecting genes to analyze further (Table 3-4), we used a combination of criteria, focusing on genes with extreme fitnesses (enhanced or reduced desiccation survival versus WT), genes in pathways that we predicted would be involved in desiccation tolerance (e.g. DNA repair), genes present in D39 and not in strain TIGR4 (because D39 survives desiccation better), and/or genes whose TIGR4 homologues had no defects in the 20 in vitro and in vivo conditions tested by Tim van Opijnen (230, 232). The rationale for this last criterion was that such genes might have evolved to play dedicated and specific roles in desiccation tolerance. We also considered that we wanted to look at genes that represent a variety of pathways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able 3-4: Genes selected for further study. Fitnesses highlighted in yellow were significant after analysis, and fitnesses in gray were nearly significant by Student t test with Bonferroni correction. Spd_0563 is a neutral locus chosen as a control. The “Insertions” columns report the number of transposon insertion sites in the gene whose reads were compiled to calculate the fitness. 80 3.2.6 Validation of desiccation changes For all genes of interest (Table 3-4), we used the splicing by overlap extension (SOE) PCR method (86) to construct individual drug marker-replacement strains with the Chloramphenicol acetyltransferase (cat) cassette containing a constitutive promoter and cat gene. The direction of transcription of cat was always the same as the gene being replaced so as to limit possible polar effects. WT S. pneumoniae D39 was transformed with these PCR products. Initially, due to our worries about polarity, we attempted to make several mutants with cat under control of the native promoter, i.e. of the genes being replaced, but many of these SOE constructs led to no transformants when plated on blood agar containing the minimal inhibitory concentration (MIC) of Chloramphenicol (Cm). As a result, we constructed all mutants with cat under control of its own promoter, Pcat (Fig. 3-3A), and reasoned that any phenotypes resulting from polarity would be identified after complementation. Certain genes that displayed a phenotype were complemented in trans by replacing the mutant strain’s neutral gene (spd_0563) with a Spec-marked SOE construct containing the complementing gene under control of its own promoter (Fig. 3-3B). The direction of transcription of the specR gene was opposite that of the complementing gene so that the Pspec promoter did not drive expression of the other gene. There were no worries about polarity in the complemented strains because the spd_0563 (degenerate transposase) locus is flanked by bi-directional terminators. 81 !" !"#$%&' ()'*&"+,&' !"#$% !&'(% &'(% #" ;!"#$<<&'('*&"+,&' -./01234'5"&678#9:' !)*+&% !"#$%&' =>?' )*+&,% !"#$% Figure 3-3: Bacterial strain construction. Knock-out mutants (A) were made via cat cassette (under control of Pcat promoter) replacement of the gene of interest (e.g. uvrA) in the WT D39 S. pneumoniae genome. Complemented strains (B) were made with a backward-reading specR marker attached to the gene of interest with its ribosome binding site (RBS) and native promoter. This construct replaced the spd_0563 neutral locus in the mutant strain (e.g. ∆uvrA::cat), which is flanked by bi-directional terminators. Dotted lines represent sites of homologous recombination. 82 Before testing our genes of interest, we confirmed that the cat cassette has no effect on desiccation tolerance by performing a 4-day competition desiccation of WT versus the neutral mutant (∆spd_0563). The median competitive index (CI) was 1 (Fig. 3-4), verifying that: 1) the neutral locus is neutral in a desiccation assay, and 2) that the presence of cat in the genome does not affect desiccation survival. To validate the desiccation changes, as measured by fitness, of the genes in Table 3-4, we performed single-strain and competition desiccations. In testing these genes, we observed three different outcomes; first were the genes whose importance in desiccation was validated. Using single-strain 4-day desiccation assays, we confirmed the defects seen in uvrA, uvrB, and uvrC mutants (p < 0.0001 for all) (Fig. 3-5). Second, some of genes did not reproduce the Tn-seq fitness results. Mutants in spd_1168/9 and spd_1836 did not survive at levels lower than WT in single-strain assays (Fig. 3-5), as we would have expected based on their low Tn-seq fitness values. This could be attributed to: 1) poor environmental conditions for assaying desiccation tolerance, as the WT survived at levels lower than expected (median of 0.11% compared to Fig. 2-2 median 4-day survival of 4.64%); 2) the single-strain desiccation not mimicking the conditions of the aggregate library desiccation in which the mutant is a minor fraction of the entire population and thus in competition; or 3) the fact that the gene is not involved in desiccation and the Tn- seq results were incorrect. In addition to testing our genes of interest, we also desiccated ∆spd_1135, encoding nth endonuclease III, to look at a mutant in a separate DNA repair pathway. This gene had a 4-day fitness of 0.534 (non-significant by t test, likely due to a low number of insertions – see Table A-1), but survived desiccation at a higher level than the ∆uvrABC mutants. 83 10 I C 1 0.1 WT vs. ! spd_0563 Figure 3-4: WT versus ∆spd_0563 4-day competition desiccation. Bacteria were recovered 4 days after desiccation of 1:1 mixture of WT (CmS) and neutral locus mutant ∆spd_0563::cat (CmR) S. pneumoniae D39. Bacteria were plated on blood agar for enumeration, after which they were replica plated onto blood agar with and without 2 µg/mL of Chloramphenicol and incubated overnight to establish the percentage of CmR R R colonies. The T4d Cm percentage was divided over the T0 Cm percentage to determine the competitive index (CI). Circles represent biological replicates, and bar shows the median. 84 10 * * * 1 l 0.1 a v i v r u S t 0.01 n e c r e P 0.001 0.0001 Limit of Detection 0.00001 WT !1135 !1168/9 !1836 !spxB !uvrA !uvrB !uvrC Strain Figure 3-5: WT and mutant survival after 4-day single-strain desiccation. Bacteria were resuspended and plated after 4 days of desiccation to determine viability. Data were pooled from multiple biological replicate experiments and outliers measured by Grubbs’ test were removed. Symbols represent biological replicates, and bars show the medians. * p < 0.0001 by Mann-Whitney U test of WT vs. mutant 85 Due to our concern that single-strain desiccations did not mimic the large-scale competition conditions of the Tn-seq, we also conducted competition desiccations on several mutant strains versus WT. As in the single-strain assays, the ∆uvrA and ∆uvrB mutants validated (data not shown). The other strains, ∆mfd, ∆mscS, ∆mscL, and ∆spd_0475, did not repeat the reduced survival of their corresponding Tn-seq transposon insertion mutants (Fig. 3-6). Again, this may be attributed to differences in the two assays – the competition desiccation pits a single mutant against WT, whereas the initial representation of a unique mutant in the Tn-seq is vastly diminished. We did not expect to see a CI different from 1 for ∆mscL, though, as it did not have a significant defect in Tn-seq (Table 3-4) and was included in our assays because we were also studying mscS. The third outcome from these experiments is that of genes that confirmed the Tn- seq results in competition desiccation, but not in single-strain desiccations. In single- strain assays, ∆sodA and spd_0185 (cardiolipin synthase) displayed WT levels of survival (data not shown). Competing against WT, however, both mutants showed significantly reduced survival (Fig. 3-6), validating their low desiccation fitnesses from the Tn-seq and indicating that single-strain and competition desiccation assays can lead to differing results. We conclude that it is difficult to recapitulate the exact circumstances under which these genes were originally identified because each mutant was massively under- represented in the Tn-seq screen. Although these assays did not confirm the involvement of every gene of interest, it is hard to determine if this is a result of them not being important in desiccation or simply a consequence of not precisely mimicking the original Tn-seq conditions. As a result, we moved forward testing all of the genes in our subsequent assays. 86 10 * 1 I C 0.1 0.01 !mfd !mscS ! mscL !sodA !0185 !0475 Strain Figure 3-6: WT versus mutant 4-day competition desiccation. Bacteria were recovered 4 days after desiccation and plated on blood agar for enumeration, after which they were replica plated onto blood agar with and without 3 µg/mL of Chloramphenicol R R and incubated overnight to establish the percentage of Cm colonies. The T4d Cm R percentage was divided over the T0 Cm percentage to determine the CI. Symbols represent biological replicates, and bars show the medians. Outliers measured by Grubbs’ test were removed. * p = 0.001 by Wilcoxon Signed Rank test 87 In attempting competition desiccations between WT and ∆spxB::cat, we found that a ∆spxB strain is out-competed in overnight growth on blood agar. Starting with a 1:1 mixture at time of plating (confirmed by plating for CFU and patching the resulting colonies onto blood agar with and without 2 µg/mL Cm), the population skewed in favor of WT (2-3 logs higher than ∆spxB) after overnight plate growth (data not shown). We believe this is due to ∆spxB’s inability to withstand exogenous H2O2 produced by WT. In a separate set of single-strain experiments, conducted under conditions of lower humidity, our ∆spxB survived at significantly higher levels than WT across 4 weeks (Fig. 3-7A). Here, the median 4-day survival for ∆spxB was 9.96%, whereas it was 0.07% in the multi-strain experiment (Fig. 3-5); this points to the inherent variability in desiccation conditions, as well as the potential for other strains to validate under lower environmental humidity. Although significantly more ∆spxB than WT cells retained viability at every time point, both strains displayed two distinct rates of decay in their survival. As seen in our earlier WT desiccations (Fig. 2-2), there was an initial large-scale extinction followed by a small subset cells persisting with a slower rate of loss (Fig. 3-7B). As ∆spxB experienced two decay rates, albeit both less extreme than WT, it therefore seems that SpxB is detrimental to surviving any length of desiccation. 88 !" * * *** * ** 100 10 1 l a v i v r u S t 0.1 n e c r e P 0.01 0.001 0.0001 2 4 7 14 21 28 Time (Days) WT !spxB #" 100 10 l a v i v r u 1 S t n e c r e 0.1 P 0.01 0.001 0 2 4 7 14 21 28 Time (Days) 89 Figure 3-7: WT and ∆spxB desiccation survival. WT (closed circles) and ∆spxB::cat (open circles) were individually desiccated, then resuspended and plated after 2 to 28 days to determine viability. Data were pooled from multiple biological replicate experiments and outliers measured by Grubbs’ test were removed. A, All data plotted, where circles represent biological replicates, and bars show the medians. B, Medians plotted on a time-scale with lines drawn to represent two survival decay rates for each strain. * p < 0.005, ** p < 0.001, *** p < 0.0005 by Mann-Whitney U test 90 3.2.7 Stress assays Based on the pathways and putative roles of certain genes in our set (Table 3-4), we designed stress assays to investigate their roles in pneumococcal physiology. These assays utilize UV light, H2O2 stress, and osmotic shock. We tested mutants representing each type of genetic pathway in all the assays. 3.2.8 Mfd, UvrA, UvrB, and UvrC are important for surviving UV-induced damage One pathway in which multiple genes were hit in our selection is DNA repair, more specifically NER. The NER genes displayed significantly reduced fitnesses at both desiccation time points (Table 3-4), indicating that they are important for surviving desiccation. In light of this, we designed a DNA damage assay using UV light (Fig. 3-8), which is known to elicit DNA damage. The bacteria were grown to mid-exponential phase in THY + Oxyrase, washed, and resuspended in PBS. This resuspension was diluted and plated to determine a starting concentration. Fifty-microliter droplets were placed onto a Parafilm square and put into the UV cross-linker (Stratagene UV Stratalinker 1800) and subjected to controlled doses (Joules) of UV-C (254 nm) light. After UV exposure, each droplet was diluted and plated to determine the surviving bacterial concentration. Initial experiments using WT to optimize the protocol showed that S. pneumoniae survived much higher UV doses when suspended in THY versus PBS (data not shown), likely because composition or color of the THY medium acted as a protective barrier for the bacteria. In PBS, there was an inverse relationship between UV 91 dose and bacterial survival, with approximately five-fold loss of viability at 5 mJ, ten- fold loss at 15 mJ, and nearly 1,000-fold loss at 50 mJ (Fig. 3-9). 92 <++)%GHI%J%K,L"03+% $(&2%3&0"&+"%461&6"+% AB%&"+0&% C(16&+%D%-10&+% 5#"%EFA% =>%?@%)"#-1+&% 4+.&"(567+8% $032% 9+363-+.)% C(16&+%D%-10&+% (.%:;<% 5#"%EFA% !"#$%&#%'()*% +,-#.+./01%-203+% Figure 3-8: Schematic of UV stress assay. A starter culture was thawed and used to seed a 3 mL THY + Oxyrase culture, which was grown to mid-exponential phase. This was centrifuged, washed, and suspended in PBS. A small volume was diluted and plated to establish the untreated (0 mJ) concentration. A 50 µL droplet was placed onto a 1 in2 Parafilm square, and this was placed into the Stratagene UV Stratalinker 1800 cross- linker for treatment with a controlled dose of UV-C (254 nm) light. After treatment, the droplet was also diluted and plated to determine the surviving bacterial concentration. 93 100 10 l a v i v r u S t n 1 e c r e P 0.1 0.01 2.5 5 7.5 10 15 20 30 40 50 UV Dose (mJ) Figure 3-9: WT survival after UV treatment. Mid-exponential phase WT S. pneumoniae D39 was centrifuged and resuspended in PBS. A sample was taken to dilute and plate for CFU (0 mJ dose), then 50 µL droplets were exposed to various UV doses (2.5 – 50 mJ), after which they were diluted and plated for CFU. Percent survival was calculated by dividing the CFU of treated samples over the initial 0 mJ CFU. Symbols represent biological replicates, and bars signify the medians. 94 When testing the mutants, we applied two doses of Joules (5 mJ and 20 mJ based on the WT results) because we wanted to know if they would display different survival trends depending on dosage. The neutral ∆spd_0563 mutant mimicked the WT result, confirming that cat does not affect survival (Fig. 3-10). Additionally, the survival of ∆sodA, ∆spxB, ∆mscS, ∆spd_1168/9, and ∆spd_1836 mirrored that of WT at both doses, which suggests that these proteins are not important for surviving or responding to UV stress or DNA damage. Conversely, the NER mutants all displayed significantly reduced survival compared to WT. The ∆mfd strain demonstrated reduced viability compared to WT – median survival of 19.9% vs. 50.5% and 1.1% vs. 5.7% at 5 mJ and 20 mJ, respectively. All three uvr mutants clustered together, with nearly 100-fold less survival than WT at 5 mJ and more than 1,000-fold less at 20 mJ. We also tested a ∆spd_1135 strain, which lacks Nth endonuclease III, an enzyme important in BER. Interestingly, it showed no difference in viability compared to WT (Fig. 3-10). It is possible that UV light and desiccation lead to similar types of DNA damage, to which the NER, but not the BER, pathway responds. 95 #" ** *** *** *** 100 10 l 1 a v i v r u S t 0.1 n e c r e P 0.01 0.001 0.0001 WT mfd ! uvrA uvrB uvrC 1135 sodA spxB mscS 0185 0475 1836 ! ! ! ! ! ! ! ! 1168/9 ! ! ! uvrB::uvrB ! 0563 (neutral) ! Strain !" * *** *** *** 100 10 l 1 a v i v r u S t 0.1 n e c r e P 0.01 0.001 0.0001 WT mfd ! uvrA uvrB uvrC 1135 sodA spxB mscS 0185 0475 1836 ! ! ! ! ! ! ! ! 1168/9 ! ! ! uvrB::uvrB ! 0563 (neutral) ! Strain 96 Figure 3-10: Bacterial survival after UV treatment. Mid-exponential phase bacteria were centrifuged and resuspended in PBS. A small volume was diluted and plated to establish the untreated (0 mJ) bacterial concentration. A 50 µL droplet was exposed to a 5 mJ (A) or 20 mJ (B) dose, after which it was diluted and plated to determine the surviving bacterial concentration. Percent survival was calculated by dividing the CFU of treated samples over the initial untreated CFU. Symbols represent biological replicates, and bars signify the medians. * p < 0.05, ** p < 0.01, *** p < 0.0001 by Kruskal-Wallis test with Dunn’s post-test on WT versus mutant strains 97 3.2.9 SpxB and SodA are important for surviving hydrogen peroxide stress Oxidative stress occurs during desiccation in other bacteria (27, 32, 52, 58, 65, 156) and, because sodA and spxB had very different survival from WT in the initial selection (Table 3-4), we subjected our strains to an H2O2 stress assay. Both SodA and SpxB are involved in oxidative stress in S. pneumoniae (Fig. 1-2). SodA, a manganese- dependent superoxide dismutase, catalyzes the redox reaction of superoxide into O2 and H2O2, and therefore contributes to a small portion of the cellular H2O2 level. SpxB is the pneumococcal pyruvate oxidase and the major producer of H2O2, yet perversely is also required to survive exposure to H2O2 (174, 183). In optimizing this assay, we found that using plate-grown bacteria instead of mid- exponential cultures eliminated confounding effects that Oxyrase added in the growth medium may have on the bacteria. We plated an individual strain as a semi-confluent lawn, which was resuspended in PBS to a mid-exponential-equivalent OD. This was centrifuged and the cell pellet subsequently resuspended in PBS with or without H2O2. Samples for CFU were taken immediately after this initial resuspension, as well as at subsequent time points over three hours (Fig. 3-11). Initial experiments testing various concentrations of H2O2 in THY established 10 mM as an appropriate dose to use in subsequent assays. There was a lag in bacterial death after approximately 1.5 hours of 10 mM H2O2 exposure, after which point WT S. pneumoniae began to die (Fig. 3-12). 98 !"#$%&'&()*&+%"",& -./0&,$#.$%.&+1"$1.%& /2%.3456$&/3&7"//8#.&& 9#,6&,%04:+/3;1%3$& D%,1,?%38& "#<3&<4$6&!=>& 43&!=>& C4"1$%&I&?"#$%&-/.&JKL& >1,?%38&#$&048:& +%3$.4-15%& #$&H):M)N&43$%.2#",& %@?/3%3A#"&BC*))& C4"1$%&I&?"#$%&-/.&JKL& D%,1,?%38&43& !=>&E&()&0F&GHBH& Figure 3-11: Schematic of H2O2 stress assay. A starter culture was thawed and diluted approximately 1:2,000 such that ~106 cells were plated onto a blood agar plate as a semi- confluent lawn. The following day, the plate was washed with PBS, suspending the bacteria to a mid-exponential phase concentration as measured by OD600. This was split into two aliquots for centrifugation and resuspension in PBS (control) and PBS + 10 mM H2O2. Small volumes were diluted and plated to establish the starting bacterial concentration. Tubes were kept at room temperature for 3 hours, with CFU being measured at 20- to 30-minute intervals. 99 100 10 1 l a v i v r u 0.1 S t n e c r e 0.01 P 0.001 0.0001 Limit of Detection 0.00001 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 Time (Min) THY THY + 10mM H2O2 THY + 50mM H2O2 THY + 1M H2O2 Figure 3-12: WT survival after exposure to varying levels of H2O2. Mid-exponential phase WT S. pneumoniae D39 was centrifuged and resuspended in THY (circles), THY + 10 mM H2O2 (squares), THY + 50 mM H2O2 (triangles), or THY + 1 M H2O2 (inverted triangles). The cultures were immediately diluted and plated for initial CFU, and samples were taken for CFU at 20- or 30-minute intervals across 3.5 hours. Percent survival was calculated by dividing time point CFU over the initial CFU. Symbols represent the mean of two experiments. 100 When we tested our mutant strains in the H2O2 assay, we found that ∆spd_0563 (neutral), ∆spd_0185, ∆spd_0475, ∆spd_1168/9, ∆1836, ∆mscS, and ∆uvrB survived at levels comparable to WT (Fig. 3-13A). The control samples, which were suspended in PBS, showed little to no loss over the three-hour time-course, and more than 10% of the 10 mM H2O2-treated samples retained viability. This suggests that none of these genes are important for surviving oxidative stress. Mutants in spxB and sodA, however, displayed significantly faster rates of killing (Fig. 3-13B), with approximately ten-fold loss by 30 and 60 minutes of H2O2 exposure, respectively. Their complemented strains, ∆spxB::spxB and ∆sodA::sodA, showed WT levels of survival (Fig. 3-13B), confirming that their H2O2 sensitivity is specifically due to the loss of either spxB or sodA. S. pneumoniae does not encode a catalase to mitigate the effects of its own endogenous H2O2,. Our H2O2 stress experiments did not include Oxyrase, an enzyme system that removes oxygen from the medium, but we were curious as to whether or not it would have an effect on the ∆spxB strain. A growth curve showed that the WT grew more slowly and to a much lower final concentration without Oxyrase, while the ∆spxB mutant had only a slight growth defect without it (Fig. 3-14). Adding two-fold more Oxyrase did not improve either strain’s growth dynamics (data not shown). This growth curve supports the idea that the increased desiccation tolerance of ∆spxB is due to its reduced H2O2 production. 101 !" PBS samples 100 WT !0563 (neutral) !0185 10 !0475 !uvrB !mscS 1 !1168/9 l a !1836 v i v r u 0.1 S t n H2O2 samples e c r 0.01 WT e P !0563 (neutral) !0185 0.001 !0475 !uvrB !mscS 0.0001 !1168/9 !1836 0.00001 0 30 60 90 120 150 180 Time (Min) #" * ** ** ** ** ** *** *** *** *** *** *** PBS samples 100 WT !0563 (neutral) !spxB 10 !sodA !spxB::spxB !sodA::sodA 1 l a v i v r H2O2 samples u 0.1 S WT t n !0563 (neutral) e c r 0.01 !spxB e P !sodA !spxB::spxB 0.001 !sodA::sodA 0.0001 0.00001 0 30 60 90 120 150 180 Time (Min) 102 Figure 3-13: Bacterial survival after H2O2 treatment. Bacterial strains were grown individually on blood plates as semi-confluent lawns, the lawn was washed with PBS to achieve a mid-exponential phase OD600, and this was washed and then resuspended in PBS (closed symbols) or PBS + 10 mM H2O2 (open symbols). Survival was measured by cell count compared to starting concentration. Symbols represent biological replicates, and bars signify the medians. Strains with survival similar to WT (black circle) and neutral ∆spd_0563::cat (blue diamond) were ∆spd_0185::cat (turquoise inverted triangle), ∆spd_0475::cat (brown diamond), ∆uvrB::cat (orange inverted triangle), ∆mscS::cat (gray circle), ∆spd_1168/9::cat (yellow square), and ∆spd_1836::cat (red hexagon) (A). Strains with survival different from WT (black circle) and neutral spd_0563::cat (blue diamond) were ∆spxB::cat (green square) and ∆sodA::cat (purple triangle), whose complemented strains ∆spxB::spxB (lime square) and ∆sodA::sodA (pink triangle) reverted to WT survival (B). Green and purple asterisks illustrate significance by Kruskal-Wallis test with Dunn’s post-test on WT versus ∆spxB::cat and ∆sodA::cat, respectively. * p < 0.05, ** p < 0.01, *** p < 0.001 103 1 0.9 0.8 0.7 0.6 0.5 0.4 0 0 6 D 0.3 O 0.2 0.1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time (Hours) No Oxyrase + Oxyrase WT WT !spxB !spxB Figure 3-14: Growth curve of WT and ∆spxB with and without Oxyrase. WT (circles) and ∆spxB::cat (squares) were grown in THY with (closed symbols) and without (open symbols) Oxyrase. The OD600 of the samples was measured over 16 hours of growth in a plate-reader. Symbols represent the median of six replicates and error bars show the range. 104 3.2.10 MscS, SpxB, SodA, and cardiolipin synthase are affected by osmotic changes In the 4-day desiccation, the ability of mscS transposon insertion mutants to survive was significantly impaired versus WT (Table 3-4). We hypothesize that osmotic shock could happen at two points in desiccation. During the drying stage, the cells may experience an osmotic upshock because they lose water and retain solutes, or during the rehydration step, the cells may undergo osmotic downshock when they are flooded with lower osmolar solution and thus need to expel solutes. This led us to develop an osmotic downshock assay, as mechanosensitive channel proteins indirectly sense and respond to changes in the osmolarity of the cellular environment. MscS (mechanosensitive channel of small conductance) opens upon turgor pressure to mediate solute efflux. In our assay, we grew the bacteria in THY + Oxyrase to mid-exponential phase, then centrifuged them and resuspended the pellet in either THY (control) or dH2O (Millipore) for downshock (Fig. 3-15). Most strains tolerated downshock as well as WT (Fig. 3-16A, D), however several mutants displayed reduced survival compared to WT (Fig. 3-16B, D). By 2 hours after resuspension, ∆mscS experienced more death due to downshock, whereas the strain complemented in trans, ∆mscS::mscS, had WT levels of survival, pointing to MscS’s role in tolerating this process. Although it did not display a fitness change in the Tn-seq (Table 3-4), the ∆mscL mutant, which is missing the large mechanosensitive channel, was also tested in the downshock assay. It did not show a reduced survival (Fig. 3-16A, D), possibly because this experiment may impose only a slight downshock. In this assay, and under desiccating conditions, there may not be enough of an osmotic change to require the presence of large channels. Note that our attempts to increase the severity of the 105 osmotic downshock – by growing bacteria in a high osmolar solution (THY with various concentrations of NaCl) or equilibrating the bacteria to high osmolar conditions (Tris + 20% sucrose) for 30 minutes prior to downshock – had no added killing effect (data not shown). Unexpectedly, the ∆spxB and ∆sodA strains were more sensitive to osmotic downshock than WT by 3 and 4 hours, respectively. A caveat to the ∆sodA result is that the THY control samples displayed a median survival of only 19% after 4 hours (Fig. 3- 16C, D), meaning that much of the loss attributed to osmotic shock may be instead due to ∆sodA poorly tolerating incubation in medium lacking Oxyrase. However, ∆sodA still demonstrated over 20-fold lower viability after four hours in dH2O than in THY (Fig. 3- 16D), suggesting that it is at a disadvantage in conditions of low osmolarity. The ∆spd_0185 mutant appears to have undergone a more immediate downshock, which is seen by plotting the dH2O-treated time point CFU divided over their correlating THY 0-hour CFU (Fig. 3-16B). The THY control and dH2O experimental samples came from the same mid-exponential culture – the culture was grown and two aliquots were separated and spun for resuspension in either THY or dH2O (Fig. 3-15). As a result, these tubes contained the same starting concentration of bacteria. By dividing the dH2O CFU over the THY starting CFU, we can look for immediate downshock. In other words, if there is a dramatic difference in CFU upon resuspension in dH2O versus THY, then cells are experiencing lethality due to downshock. The ∆spd_0185 strain showed a median five-fold viability loss upon resuspension in dH2O versus THY, the result of near- instant downshock. Like ∆mscS, this mutant showed a further downward trend in 106 survival after 2 hours of downshock (compare the 0- and 1-hour CFU versus subsequent time points). 107 8++)%9:;%<%=,>"03+% B#.&"#1% $(&2%3&0"&+"%461&6"+% ?+363-+.)% (.%9:;% C(16&+%D%-10&+%5#"%BEF% 4+.&"(567+% 0&%GHI%(.&+"J013% C(16&+%D%-10&+%5#"%BEF% !"#$%&#%'()*% ?+363-+.)% +,-#.+./01%-203+% (.%):@=% =3'#/4%)#$.32#4A% Figure 3-15: Schematic of osmotic downshock assay. A starter culture was thawed and used to seed a 3 mL THY + Oxyrase culture, which was grown to mid-exponential phase. This was split into two aliquots for centrifugation and resuspension – one in THY (control) and the other in dH2O (downshock). Samples were immediately diluted and plated to determine starting concentration. Tubes were kept at room temperature, and this process was repeated at 30-minute intervals over 4 hours. 108 !" 100 l a v i v 10 r u S t n e c r e P 1 0.1 0 1 2 3 4 Time (Hours) WT !spd_0475 !spd_1836 !spd_0563 (neutral) !spd_1168/9 !mscL !uvrB * #" * ** ** ** ** *** *** ** *** ** 100 l a v i v 10 r u S t n e c r e P 1 0.1 0 1 2 3 4 Time (Hours) WT !mscS !spd_0185 !spd_0563 (neutral) !mscS::mscS !spxB !sodA 109 !" 100 l a v i v 10 r u S t n e c r e P 1 0.1 0 1 2 3 4 Time (Hours) WT !spd_1168/9 !uvrB !spd_0185 !spd_0563 (neutral) !spd_1836 !mscS !spxB !spd_0475 !mscL !mscS::mscS !sodA Figure 3-16: Bacterial survival after osmotic downshock. Mid-exponential phase bacteria were centrifuged and resuspended in THY or dH2O. A small volume was immediately diluted and plated to establish the starting bacterial concentration. Percent survival was calculated by dividing the CFU of dH2O samples over the initial THY (control) CFU. Symbols represent biological replicates, and bars signify the medians. A, Strains with survival similar to WT (closed circle) and neutral ∆spd_0563::cat (open circle) were ∆spd_0475::cat (closed triangle), ∆spd_1168/9::cat (closed inverted triangle), ∆spd_1836::cat (closed diamond), ∆mscL::cat (star), and ∆uvrB::cat (x). B, Strains with survival different from WT (closed circle) and neutral ∆spd_0563::cat (open circle) were ∆spd_0185::cat (open triangle), ∆spxB::cat (open inverted triangle), ∆sodA::cat (open diamond), and ∆mscS::cat (closed square), whose complemented strain ∆mscS::mscS (open square) reverted to WT survival. C, THY control sample medians. D, Quantification of fold loss in survival measured by comparing the median of 4-hour THY control samples versus 4-hour dH2O downshock samples. Data are presented in order of increasing fold loss (highlighted in red). * p < 0.05, ** p < 0.01, *** p < 0.0001 by Kruskal-Wallis test with Dunn’s post-test on WT versus mutants 110 Because it is possible that the bacteria undergo a slow osmotic upshock during the drying phase of desiccation, we also performed an osmotic upshock assay. This protocol involved growing the bacteria to mid-exponential phase in THY + Oxyrase, then incubation in dH2O for 30 minutes to equilibrate them to a low osmolar solution before centrifuging and resuspending them in either dH2O (control) or high osmolar solution Tris + 20% sucrose (upshock). We tested WT, ∆mscS, and ∆mscL in this assay, and saw that all three strains survived at nearly 100% over the 3-hour experiment (Fig. 3-17, closed bars). Therefore, no upshock stress appeared to occur. Instead, ∆mscS showed the same reduced survival by 2 and 3 hours after resuspension in dH2O as seen in the downshock assay (Fig. 3-16B). (The 2- and 3-hour time points for the dH2O samples actually represent 2.5 and 3.5 hours in dH2O when the 30-minute incubation at the start of the experiment is taken into account.) From this assay, we conclude that it is more likely that the cells undergo downshock during rehydration from desiccation than upshock during drying. 111 20% sucrose (upshock) 100 WT !mscL !mscS l a v i v r 10 dH O (control) u 2 S t WT n e c !mscL r e !mscS P 1 0.1 0 0 0 0 0 0 120 180 120 180 120 180 120 180 120 180 120 180 WT !mscL !mscS Time (Min) Figure 3-17: Bacterial survival after osmotic upshock. Mid-exponential phase bacteria grown in THY + Oxyrase were centrifuged and suspended in dH2O for a 30- minute equilibration, after which this culture was split into two aliquots for centrifugation and resuspension – one in dH2O (control) and the other in Tris + 20% sucrose (upshock). A small volume was immediately diluted and plated to establish the starting bacterial concentration. Tubes were kept a room temperature, and CFU was measured at 2- and 3- hours after resuspension. Percent survival was calculated by dividing the time point CFU over the starting CFU. WT (black), ∆mscL::cat (pink), and ∆mscS::cat (red) strains were tested in control (open bars) and upshock (closed bars) conditions. Bars represent the mean of two experiments, and error bars signify the standard error of the mean (SEM). 112 113 Table 3-5. Compiled results from desiccation and stress assays. Listed here are results of our genes of interest from the desiccation and stress assays. The 2-day and 4- day fitnesses are from the Tn-seq desiccations. Fitnesses highlighted in yellow were significant after analysis, and fitnesses in gray were nearly significant by Student t test with Bonferroni correction. The “4-day Desiccations” columns include data calculated from the validation desiccations (Section 3.2.6). The “Single-Strain” column values are presented as the fold-difference in 4-day desiccation survival between the mutant median and the WT median. The “Competition” column values are the CI’s from 4-day desiccations. The “Stress Assays” columns include data from the three stress assays (Sections 3.2.8, 3.2.9, and 3.2.10). The “UV” column values are presented as the fold- difference in survival between the mutant median and the WT median after 5 mJ UV exposure. The “H2O2” column values are presented as the fold-difference in survival between the mutant median and the WT median after 3 hours of 10 mM H2O2 treatment. The “Downshock” column values are presented as the fold-difference in survival between the mutant median and the WT median after 4 hours of osmotic downshock from THY to dH2O. 114 3.3 Discussion After analyzing our 2- and 4-day Tn-seq desiccations, we selected a set of genes from various pathways for further study. These genes encode proteins involved in DNA repair, cell wall/membrane maintenance, oxidative stress response, response to osmotic changes, and one of unknown function (Table 3-4, Table 3-5). In single-strain desiccation assays, we were able to validate the reduced tolerance to desiccation seen in uvrA, uvrB, and uvrC mutants, as well as the enhanced desiccation tolerance of spxB mutants. Competition desiccations also validated the reduced tolerance of sodA and cardiolipin synthase mutants. The other genes did not replicate the Tn-seq results, and we attribute this mostly to the experimental conditions being different – namely that the mutants in the Tn-seq were a tiny proportion of the overall population, which may have predisposed them to be stressed upon entering desiccation, unlike in single-strain desiccations. It would be interesting to construct a fully saturated transposon insertion library and to combine multiple technical replicates for the Tn-seq desiccation to eliminate the chance of stochastic loss. Combining this method with comparing multiple biological replicates may allow for the identification of additional genes involved in pneumococcal desiccation tolerance, and may enable time points past 4 days to be analyzed. Using stress assays, we investigated the roles of our selected genes in pneumococcal physiology to better understand how this plays into desiccation tolerance. First, we used UV light to induce DNA damage and found that Mfd and the UvrABC complex all play a role in S. pneumoniae’s DNA damage repair. This repair is likely a key process needed to emerge from desiccation, which has been shown to induce DNA 115 damage in other bacteria, such as E. coli (9, 54, 179). The significantly decreased survival of the uvr mutants confirms their proteins’ role in DNA repair, something that had been assumed based on bacterial homologues, but had not been demonstrated in S. pneumoniae. UvrC was previously identified in a screen as required for S. meliloti Rm1021 desiccation survival, and subsequent experiments also showed the importance of UvrA and UvrB in this process (87). Mfd seems to have a lesser role in surviving DNA damage than the Uvr proteins, likely because it is involved in transcription-coupled repair. The time of UV exposure is brief (on the order of several seconds), and therefore transcription-coupled repair does not seem to be the main mechanism by which UV- induced DNA repair occurs. Because Mfd and UvrABC are the proteins needed for NER (224), we wondered if other DNA repair pathways would show similar survival trends. Of the pneumococcal DNA repair pathways, only the NER genes showed a reduced ability to survive desiccation as measured by our Tn-seq. Most recombinational repair genes are essential, and therefore cannot be tested, but genes involved in direct repair, BER, and MMR showed no fitness significantly different from the neutral value of 1 (Table A-1). To further test this, we made a mutant lacking nth endonuclease III (spd_1135), which survived both desiccation and UV stress at WT levels. We therefore hypothesize that NER is the critical DNA repair pathway for pneumococcal desiccation tolerance. Also essential to survival is the response to oxidative stress. Oxidative stress is known to play a role in the desiccation survival of other bacteria, including S. aureus, A. baumanii, and E. coli (32, 58, 179). In particular, Gayoso and colleagues identified superoxide dismutase as important in A. baumanii desiccation tolerance (58). We found 116 that ∆spxB and ∆sodA strains have significantly faster rates of killing upon H2O2 exposure than WT or the other strains tested. The ∆spxB result is not novel, but does confirm previously published findings that spxB mutants show increased killing versus WT upon addition of exogenous H2O2 (174). Reduced survival of ∆sodA has been shown in S. pneumoniae upon exposure to paraquat (253), which produces damaging superoxide radicals. Our work is the first to test the effect of exogenous H2O2 on a ∆sodA strain, and suggests that increased H2O2 can contribute to increased superoxide, requiring SodA to mediate its effects. SpxB and SodA likely have opposite roles in desiccation tolerance, with SpxB contributing to the accumulation of potentially damaging levels of ROS, and SodA helping to eliminate them. This is reflected in the fact that spxB mutants showed enhanced desiccation tolerance in the Tn-seq and individual desiccation experiments, while sodA mutants displayed reduced fitness in the Tn-seq and in competition desiccations. Although we hypothesize ROS contribute to DNA damage during desiccation, ∆uvrB did not show enhanced susceptibility to H2O2. In line with this, E. coli ∆sodA∆sodB mutants are sensitive to H2O2, while ∆uvrABCD mutants are not (91). E. coli undergoes two types of ROS-mediated killing – mode 1 involves DNA damage and requires active metabolism during exposure, whereas mode 2 causes uncharacterized damage and occurs in the absence of metabolism (70, 90, 91). As E. coli ∆sodA∆sodB was sensitive to mode 1 killing (91), S. pneumoniae ∆sodA also may be affected by mode 1 killing. It seems likely, then, that S. pneumoniae SodA would be necessary during the drying phase when it can scavenge ROS that accumulate as a result of still-active metabolic processes. 117 Perhaps unexpectedly, other genes involved in detoxification of and protection from ROS did not come out of the Tn-seq data. Some genes, like the psa operon, had no transposon insertions in our aggregate library, and thus no fitness was calculated. The ciaRH genes had reduced fitnesses in the 4-day Tn-seq (fitnesses = 0.508 and 0.614), yet with only two transposon insertions per gene, there was not enough coverage for a robust fitness calculation. Therefore we cannot eliminate the possibility that these genes are involved in desiccation tolerance, a possibility that could be tested with a more saturated transposon insertion library or by desiccating individual deletion mutants. Other genes, like rgg (2-day fitness = 0.942, 4-day fitness = 1.121) and the CTM complex, had a sufficient number of transposon insertions, but simply did not have fitnesses significantly different than the WT value of 1. Rgg, a repressor up-regulated during aerobic growth (20), does not seem to be activated during desiccation. Even through the cells are under aerobic conditions when desiccated, they are not growing and thus would not need Rgg to be functioning. The CTM complex is comprised of two electron-transferring CcdA proteins, two thioredoxin-like Etrx proteins, and methionine sulfoxide reductase MsrAB2 (6, 196). Electrons move from cytoplasmic thioredoxin TrxA to CcdA1, TlpA, and finally to MsrAB2, which donates them to oxidized surface proteins. Recent work has shown that D39 mutants lacking either MsrAB2 or both Etrx proteins experience higher rates of killing upon exposure to exogenous H2O2 (196). Thus, we are not surprised that ccdA-1, ccdA-2, etrx1, and etrx2, each which have a functionally redundant homologue, did not have fitnesses different from WT in our Tn-seq. MsrAB2, which we might expect to have reduced fitness, instead had a 2-day fitness of 0.884 and a 4-day fitness of 0.986. This may be because MsrAB2 rescues surface proteins whose methionine and 118 cysteine residues can be oxidized by extracellular ROS (6, 196). These surface proteins may not be the targets of most desiccation damage, and we hypothesize that most desiccation-induced ROS are produced endogenously, either by SpxB or as byproducts of other metabolic activities. ROS are also known to have effects on the bacterial membrane (67), and the ability to change membrane composition is an important survival strategy used to adjust to changing environmental conditions (8, 175). When switched to a low osmolar solution in our osmotic shock assay, ∆mscS, ∆sodA, ∆spxB, and ∆spd_0185 (cardiolipin synthase) all had reduced survival compared to WT. MscS, SodA, and SpxB seem to be important for long-term survival in low osmolar conditions, while cardiolipin synthase is helpful in surviving immediate downshock. In other bacteria, osmotic downshock is measured to occur within several seconds (185), not over several hours; therefore it may be more reasonable to refer to our downshock assay as a “low osmotolerance” assay. While mscS mutants showed reduced desiccation fitness in the Tn-seq, there was no immediate loss in ∆mscS’s viability upon suspension in dH2O. This could be because the assay does not precisely reproduce the osmotic changes that occur during desiccation, or because MscL may act in place of the missing MscS and thus only a double mutant would display reduced survival upon immediate downshock. In addition to opening and closing mechanosensitive channels in response to changing conditions, bacteria can alter their membrane fatty acid composition in response to ROS, including H2O2. Previous work showed that S. pneumoniae D39 has reduced cardiolipin content in its membrane under anaerobic versus aerobic conditions (175), suggesting that cardiolipin synthase is more active in oxygenated environments. A D39 119 ∆spxB mutant also displays reduced cardiolipin levels (175). Differences in membrane fatty acid content between WT and ∆spxB can be explained via SpxB-produced H2O2 oxidation of the active site cysteine-thiol of FabF, a critical factor in regulating fatty acid composition (15). B. subtilis cardiolipin and PG synthases are osmotically regulated (130, 190), and under osmotic stress, E. coli cardiolipin levels increase via induction of cardiolipin synthase (79). It appears that S. pneumoniae cardiolipin synthase also is osmotically regulated. Because both ∆spd_0185 and ∆spxB showed dramatically reduced 4-hour survival in dH2O, surviving downshock (or at least tolerance to low osmolar conditions in the case of ∆spxB) may require either an increase in cardiolipin content or sufficient pre-existing cardiolipin present to adapt to this condition. Testing strains containing these genes complemented in trans would confirm that the effects we saw are solely due to the gene deletion. It is also possible that SpxB and SodA, whose mutants were sensitive to low osmolarity by 3 and 4 hours, respectively, are part of a larger stress regulon. Without them, the bacteria may be at a disadvantage when presented with certain non-oxidative stresses. Perhaps H2O2 acts as a signaling molecule or molecular switch (e.g. via reversible disulfide bond formation, or redox changes of transcriptional regulators), as has been reported in other systems (59, 128, 175, 187, 188). Like with FabF, other thiol- containing proteins use H2O2 as a mediator to regulate activity (187, 188). By not making H2O2, a ∆spxB strain would not have a signal crucial for protein regulation, whereas ∆sodA may not be able to detoxify ROS fast enough, thus impeding a putative signal cascade or critical protein modification. 120 Overall, S. pneumoniae’s desiccation tolerance is linked to its ability to resist stresses, including those resulting from osmotic changes and ROS, which can lead to DNA and possibly protein damage seen in other bacteria (52, 179). These stresses can occur at various points during desiccation – in the drying process, while dry, and during rehydration. We hypothesize that ROS byproducts of SpxB and metabolic processes accumulate and cause damage to DNA – and maybe cytoplasmic proteins – leading up to and during desiccation. SodA is helpful in neutralizing certain oxygen species before they can cause damage, while SpxB contributes to higher ROS levels. ROS levels, MscS, and cardiolipin synthase may help S. pneumoniae to deal with osmotic shock and necessary membrane changes faced by the bacteria upon rehydration after desiccation. In coming out of this dry state and restarting metabolism and transcription, there is ROS- induced damage to fix. This is when Mfd and the UvrABC NER complex are needed to repair DNA damage. 121 CHAPTER 4: Summary and Future Perspectives 122 4.1 Summary The medical community has long considered the pathogen S. pneumoniae to be transmitted person-to-person exclusively via aerosolized respiratory droplets. While a limited number of studies have investigated its ability to survive in the environment under various conditions (149), and even one report suggested the likelihood of fomites serving as the infectious source of an outbreak (83), this is the first body of work to thoroughly investigate the desiccation tolerance phenotype of the pneumococcus. Using a novel desiccation assay, in which plate-grown bacteria are spread onto polystyrene plates, we established that S. pneumoniae serotype 2 strain D39 retains viability for at least 28 days under ambient conditions. Since the publication of our initial experiments, other groups have demonstrated survival of S. pneumoniae after 1 to 30 days of desiccation (138, 227). We elected not to control the temperature and relative humidity, wanting instead to assay year-round pneumococcal survival under similar conditions to those putatively experienced by the pathogen during transmission in the environment. As S. pneumoniae was able to survive at room temperature (~23 to 25˚C) under varying low- to mid-level humidities (~18 to 45%) for four weeks, this confirms that pneumococci can tolerate extended periods outside of the human nasopharynx and supports the hypothesis that these surviving bacteria may serve as an alternate source of transmission from fomites to new human hosts. The pneumococcus rapidly lost viability over the first four days of desiccation (< 1% survival), after which time the rate of death slowed, leaving a population that persists long-term. Subsequent experiments showed that survival and starvation are separate phenomena, suggesting that different stress regulons are involved in bacterial survival of 123 these stresses. Additionally, capsule was dispensable for serotype 2 survival over a week of desiccation. As capsule is the defining factor between serotypes, we tested 17 strains, representing 14 unique serotypes, and were able to recover all strains after two days of desiccation. Moreover, there was no clear pattern in survival among the three strains in which both opaque and transparent variants were tested, again supporting our conclusion that capsule is not required for pneumococcal desiccation tolerance. Persistence of the varying strains tested suggests that this ability is intrinsic to the core S. pneumoniae genome. To evaluate these underlying genetic mechanisms, we employed the Tn-seq method to probe the pneumococcal genome for genes involved in desiccation tolerance. Several pathways – DNA repair, oxidative stress response, osmotic stress response – were identified and further investigated. Using stress assays (Table 3-5), we established that Mfd and the UvrABC excinuclease are important in pneumococcal DNA repair, and that SpxB and SodA are essential for surviving exogenous H2O2 exposure. SpxB, SodA, and MscS were needed for long-term tolerance in medium with low osmolarity, while cardiolipin synthase helped cells undergoing immediate osmotic downshock. Therefore, various stress responses are crucial for S. pneumoniae to tolerate numerous aspects of the desiccation process. We hypothesize that during drying, SpxB contributes to a cellular pool of ROS, which damage DNA, while SodA detoxifies highly reactive superoxide. Upon rehydration, cells experience a dramatic osmotic shift, requiring MscS and cardiolipin synthase to respond to turgor pressure and associated membrane changes. Finally, cells also are faced with DNA damage, which is identified and repaired by Mfd and UvrABC. In all, this body of work advances our understanding of S. pneumoniae 124 desiccation tolerance and the genes involved in recovery, and suggests that the traditional view of transmission should be amended to include the alternative fomite pathway (Figure 4-1). It represents an important first step toward identifying pneumococcal desiccationsurvival mechanisms, which will enable medical professionals to make informed decisions about pneumococcal transmission and improved fomite decontamination practices. 125 !"#$%#&'()#%*+',-( .%&,39"&3,')(:'3+,:23#'( 56-2'*4$+'0&%&72%+%&"/')( .%+%&"/')0"&1'2,')(*34'&,( ;%#<'#-=(:3&)-(( *34'&,( 8%9",'-( Figure 4-1: Pneumoccocal transmission routes. Colonized or infected persons can transmit S. pneumoniae via aerosolized respiratory droplets, which can then colonize previously non-colonized individuals. Alternatively, these airborne droplets can settle on and contaminate a fomite, or the infected person’s hands can spread pneumococci onto a fomite, a site which non-colonized individuals can subsequently contact and thereby transfer bacteria to themselves. In hospital settings (dotted lines), healthcare workers can function as unwitting intermediaries in transmission – either from infected to susceptible patient, or by contacting contaminated fomites and transferring bacteria from there. Adapted from reference (169). 126 4.2 Future Perspectives 4.2.1 Broader experiments It would be informative to conduct additional experiments to advance our understanding of pneumococcal desiccation tolerance, as plenty of questions still remain. For example, does rehydration and plating for CFU accurately represent the surviving, transmissible bacterial population? Some bacteria are known to shift to a “viable-but- non-culturable” (VBNC) state (164, 180, 237), and Salmonella spp. were shown to enter this VBNC state during desiccation (65). To eliminate the possibility that plating for CFU underestimates the number of surviving cells that could be infectious, live/dead staining could be conducted on the resuspended bacteria to confirm equivalent results via both methods. This brings up the question of whether or not S. pneumoniae enters a state of metabolic stasis or dormancy upon desiccation. Maybe a small portion of the population undergoes such a transition, and these are the cells capable of surviving long- term on fomites. This could be clarified with transcriptional fusions to housekeeping genes. By spreading bacteria containing a fluorescent reporter fusion onto glass microscope slides, the cells on these fomites could be viewed under the microscope and the relative fluorescence quantified. Taking this one step further, fusions could be made with various genes identified in our Tn-seq screen to definitively establish during which stage of the desiccation process – desiccation, while dry, or rehydration – they are maximally expressed. It theoretically is possible that excitation with the laser could affect desiccation tolerance, in which case a luciferase fusion may be an alternative. By tracking the light emitted over the entire desiccation period, gene activity at each stage in the process could be ascertained. 127 While we desiccated only S. pneumoniae on fomites, in situations of person-to- fomite transmission, the bacteria would be surrounded by a matrix of other bacterial species, mucus and/or saliva, being that the most likely route of dissemination is through coughing, sneezing, and nose-blowing. Consequently, mixed-species desiccations should be conducted. S. pneumoniae is known to form mixed-community biofilms in the nasopharynx, and environmental changes resulting from interactions with other pathogens not only can disperse pneumococci from the biofilm, but also can change their gene expression profile and virulence phenotype (135). It is possible that the presence of other bacteria and bacteriophages will have a marked effect on S. pneumoniae’s survival. Mucus and saliva may act as an insulating barrier between the pneumococcus and the ambient environment, either protecting the cells from true desiccation and creating micro- environments on the fomite or providing the cells with additional nutrients to use for survival or during rehydration. This idea is supported by a study that found pneumococci survive for 24 hours significantly better in concentrations of over 25% human saliva mixed with PBS as compared to 0-25% saliva; in fact the S. pneumoniae in > 25% saliva mixtures grew during this time, suggesting they can feed on components in saliva (233). The authors’ genome-wide screen identified specific metabolism and transport genes to be essential for pneumococcal survival in saliva (233), however they did not conduct experiments under desiccating conditions. Using their protocol, saliva could be collected and sterilized prior to mixing with S. pneumoniae for desiccation. Similarly, the same experiment could be attempted with sterilized mucus, but since mucus is generally more viscous than saliva and this could hamper filtering, a related assay could use varying 128 concentrations of industrially available mucin. These experiments would more closely recapitulate real-world transmission and environmental survival scenarios. Like saliva and mucus, biofilms provide a protective matrix for bacteria to survive desiccation. Bacteria in biofilms often survive insults that planktonic cells do not, so biofilms may be the reason why vegetative cells can survive desiccation for prolonged periods of time (169). Work on S. pneumoniae has shown that biofilm-derived cells survive desiccation at dramatically higher levels than non-biofilm cells (138), and separate research identified MRSA in biofilm form on five of six assayed hospital surfaces post-disinfection with bleach (169). Pneumococcal biofilms, with their extracellular sugars, proteins, and DNA, may enable cells to better survive the many stresses experienced during desiccation. It is unclear whether or not this protective extracellular matrix (ECM) was present in the experiments testing S. pneumoniae biofilm-cell survival on fomites. The bacteria were washed once with PBS and centrifuged prior to being resuspended in PBS and dried down on the fomite (138). This process might not separate all of the ECM components from the bacterial cells, meaning that this protective architecture may have contributed to the higher levels of survival achieved by the biofilm-derived bacteria. Alternatively, cells in biofilms may be better able to tolerate desiccation as a result of different signaling pathways initiated in the biofilm versus in planktonic cells. Biofilms contain cells at a variety of metabolic states (215) and it is possible that some of these micro-communities may be better prepared for desiccation. 129 4.2.2 Genetic pathways While our transposon insertion library, with 28,000 mutants, was nearly saturated for the D39 genome, it would be beneficial to employ Tn-seq on a completely saturated library. This would better guarantee coverage of all genes, no matter how small, and would ensure transposon hops into many sites within genes. With its 28,000 unique transposon insertions, our library has a transposon every 73 nucleotides on average (assuming even distribution throughout the genome). S. pneumoniae is a low-GC-content bacterium (220), so the TA-site-specific Magellan6 transposon is a suitable choice for additional library preparation. A fully saturated library, in addition to multiple desiccated replicates per time point, and potentially a higher starting concentration of desiccated bacteria (thus leading to a higher total number of surviving cells), should allow for a more thorough interrogation of the genome. We could then more conclusively determine whether or not additional genes, such as other oxidative stress genes, are involved in desiccation tolerance. Our lab also has saturated libraries in the TIGR4 background, which have been evaluated in a variety of conditions (230, 232), but not in desiccation. It would be interesting to make libraries in even more serotypes and assess the importance of various genes in each background. For example, one of the mutants we tested, ∆spd_1168/9, is a two-gene deletion of permeases in an ABC transporter. We chose to include this mutant in our stress tests because multiple genes in this operon were hit in the 2- and 4-day Tn-seq desiccations, but also because this operon is encoded by D39 and not TIGR4. Other than this transporter, most of the genes in this category (found in D39, but not TIGR4) encode proteins of hypothetical function (Table A-2). While we did not see a phenotype in any of these assays, follow-up experiments could be designed to 130 determine the substrate for this ABC transporter, a step toward understanding how it is involved in desiccation tolerance. Another category of genes that warrants further investigation are those that responded to desiccation, but were unresponsive in the 20 in vitro and in vivo conditions tested by Tim van Opijnen (230, 232), as these are more likely to have a dedicated role in desiccation resistance. Genes encoding several other transporters fall into this group, in addition to cell wall and cell surface proteins, numerous hypothetical proteins, and two that we tested – spd_0475 and MscS (Table A- 3). Although the ∆mscS strain had a phenotype in the osmotic stress assay, it was not sensitive to immediate downshock. Nor was ∆mscL sensitive to osmotic stress in our assays, a somewhat surprising result. One reason for this may be that our assay does not impose a high enough stress on the cells to require opening of the large channel, something that also may not be needed during desiccation, as mscL had no fitness difference versus WT in the Tn-seq. In the annotated D39 and TIGR4 genomes, only mscS and mscL are designated as mechanosensitive channel genes. On the other hand, E. coli also has a mini channel and a K+-specific channel (115), while B. subtilis has several mscS homologues (85). Preliminary nucleotide and protein BLAST searches comparing these sequences against S. pneumoniae’s genome did not identify any candidate pneumococcal genes that may encode an additional mechanosensitive channel. It is likely that these are the only two such channels present in S. pneumoniae, making it similar to other Gram-positive bacteria whose genomes were comparably queried (131). Assuming these are the only mechanosensitive channels present, making and testing a 131 double mutant in osmotic stress assays and in desiccations would clarify if single deletion strains were not as sick due to functional redundancy of these two proteins. Instead of deleting the mechanosensitive channels, an alternative would be to use gadolinium ions (Gd3+), which are known to block these channels. By inhibiting their ability to release solutes under turgor pressure, we expect that these strains would show reduced survival under hypoosmotic stress. Gd3+ effectively blocked E. coli MscL only when the membrane contained negatively charged phospholipid phosphatidylserine, a PE precursor (49). This further points to the interconnected role of mechanosensitive channels and the lipid architecture of the membrane. Cardiolipin, too, is a negatively charged phospholipid, and may directly interact with these channels. Alternatively, its effect may be indirect; cardiolipin is involved in shaping the curvature of the membrane (146), and so its presence could have a profound effect on membrane shape under turgor pressure. Given a cardiolipin-deficient membrane, its fluidity and elasticity likely would be significantly changed, thus affecting the function of stretch-activated MscS and MscL. To confirm that a reduced ability to withstand osmotic downshock is due partially to a lack of cardiolipin, a complemented strain should be tested to see if this restores WT levels of survival under hypoosmotic stress. 4.2.3 Fomite transmission Our hypothesis regarding fomites as environmental reservoirs, and thus potential sources of pneumococcal infection, would be strengthened with experimental demonstration of fomite transmission. S. pneumoniae has been shown to survive on human hands for three hours (74, 138) and has been isolated from fomites surveyed in day care centers (123, 138). Although fomites are believed to have been sources of 132 pneumococcal infections (83), to date, there have been no reports proving S. pneumoniae transits from fomite to nasopharyngeal niche. We attempted to develop a fomite transmission assay wherein we spread and desiccated pneumococci onto sterile, compact cotton bedding squares given to mice, but we were unable to definitively establish nasopharyngeal colonization. In light of recent work on experimental human pneumococcal carriage (64), a more appropriate method would be to contaminate fomites with controlled doses of S. pneumoniae and then determine if fomite-to-hand-to- nasopharynx transmission can occur. Marking the desiccated bacteria with a drugR cassette would allow for differentiation from any pneumococci that may already be colonizing the volunteers. It also may be possible to use a tissue culture model of the human nasopharynx to look at colonization and transmission. One potential conclusion from our assays is that S. pneumoniae desiccation tolerance does not depend on RH because we were able to recover bacteria under a range of humidities (~18 to 45%). We caution against this interpretation, however, as we did observe that certain times of year – particularly late winter – were more conducive to higher levels of recovery. This increased recovery cannot be entirely attributed to low RH, but it is interesting that late winter is the same time as the annual spike in pneumococcal infections (44, 63, 111, 243). Another trend in late winter is decreased daylight hours, and thus reduced UV exposure. It is possible that this, combined with cold temperatures, forces people into close quarters and leads to increased person-to- person spread. It is also conceivable that reduced UV rays allow for longer environmental survival of disseminated bacteria, a concept supported by the low desiccation tolerance of the uvrABC-deficient mutants and their higher levels of UV- 133 induced killing in our assays. Further molecular and epidemiological studies are needed to better establish causality between any of these seasonal characteristics and both bacterial persistence in the environment and transmission between people. 4.3 Conclusion Building on earlier work on environmental survival of the pneumococcus (149), we have demonstrated long-term desiccation tolerance of S. pneumoniae. It seems very plausible that nasopharyngeally-colonizing S. pneumoniae, likely in biofilm form (138), would be expelled via normal respiratory activities and come to rest on various environmental surfaces. We have shown that large numbers of bacteria can persist for several weeks on these fomites, making them a hidden risk for pneumococcal transmission and colonization, which can occur with as little as 104 bacteria (64). Similar to other bacteria, stress responses are critical to S. pneumoniae’s desiccation tolerance. Specifically, the bacteria experience both oxidative stress and osmotic stresses, and the cells need DNA repair – particularly NER – to overcome desiccation-induced damage. Based on the responses of ∆spxB and ∆sodA to hypoosmotic stress, it is also clear that these stress responses may overlap, and that diverse genes may be part of an overall desiccation-related regulon. 134 CHAPTER 5: Materials and Methods 135 5.1 Bacterial strains 5.1.1 Strains and growth conditions The S. pneumoniae strains used in this study are listed in Table 5-1. Mid- exponential-growth-phase S. pneumoniae bacteria (grown to optical density 600 nm [OD600] 0.4-0.6 in Todd-Hewitt [Becton, Dickinson, Co.] broth supplemented with 0.5% yeast extract [Fisher] [THY] + Oxyrase [Oxyrase, Inc.]) were stored as “starter cultures” in microfuge tubes at -80˚C in THY + 15% glycerol in 0.25 to 0.5 mL aliquots. Unless otherwise noted, strains were grown in THY + Oxyrase. 5.1.2 Strain construction The S. pneumoniae strains used in this study are listed in Table 5-1, and the primers for strain construction are listed in Table 5-2. Mutant strains were created using the splicing by overlap extension (SOE) PCR method (86), followed by allelic exchange. In all gene replacement strains, the upstream (“up”) and downstream (“down”) arms of homology flanking the site of exchange were PCR amplified from D39 gDNA. The R drug cassette (cat with its promoter Pcat or spec with its promoter Pspec) was PCR amplified from a plasmid. Using step-wise SOE PCR, the up, drugR, and down arms were spliced together to create a final PCR product, which was gel purified and used to transform S. pneumoniae D39 (AC1770). Mutants were confirmed by sequencing. The complementation strains used allelic exchange at the spd_0563 neutral locus in the mutant background. The “up” and “down” arms, in addition to the gene being complemented in trans (under control of its native promoter), were PCR amplified from D39 gDNA. The “up” arm was spliced to backward-facing spec with Pspec, which in turn was spliced to the putative ribosome binding site (RBS) of the forward-facing gene, and 136 this was spliced to the “down arm.” This final PCR product was used to transform the corresponding mutant strain. 5.2 Desiccation assays 5.2.1 Desiccation An S. pneumoniae starter culture containing ~107-108 CFU was thawed, the entire contents were plated on Trypticase soy agar plus 5% sheep’s blood (Northeast Laboratory) (blood agar), and after approximately 16 hours of growth at 37˚C in a 5% CO2 incubator, the bacterial lawn was scraped off the plate surface and pooled in the center of the plate using a plastic straight-edge (catalog no. 165-3320, Bio-Rad Gel Releaser). Unless otherwise noted, the bacteria were then evenly split by eye into 3 to 4 pools, each of which was spread thinly and evenly using the plastic straight-edge onto the inner side of a 100 mm by 15 mm polystyrene petri dish lid (Fisher). There were approximately 108 CFU spread on each lid. Immediately after spreading, the bacteria on one lid were resuspended with 1.5 mL THY, serially diluted in THY, and plated to determine the number of CFU, constituting the initial time point (T0). The remaining lids were closed over the petri dish bottoms and placed in the dark at room temperature for specific times (1 hour to 28 days) until being similarly resuspended and serially diluted to determine the number of CFU. Desiccations were conducted in the dark based on previous research implicating photo-oxidative damage as a cause of enzyme, protein, and DNA damage during desiccations under light conditions. The percent remaining viable bacteria was calculated by dividing the time point number of CFU by the T0 number of CFU. Ambient temperature and relative humidity were monitored throughout the 137 duration of each experiment. The range for all experiments was 21 to 25˚C and 18 to 45% relative humidity. 5.2.2 Mid-exponential-growth-phase desiccation An S. pneumoniae starter culture containing ~107 CFU was thawed and used to seed a 10 mL culture of THY + Oxyrase. After 4 hours of growth to mid-exponential phase (OD600 0.4) at 37˚C in a 5% CO2 incubator, the culture was split evenly into 20 aliquots (480 µL each) that were spun for 7 minutes at 1500 x g at room temperature. The supernatants were removed except for 50 µL and each pellet was resuspended and transferred to the inner side of a 100 mm by 15 mm polystyrene petri dish lid, onto which it was spread thinly and evenly using the plastic straight-edge. Approximately 2 x 107 CFU were spread on each lid. Immediately after the pellet was spread, the petri dishes were dried open in the biosafety cabinet with airflow on for 30 minutes. Next, the bacteria on the four lids were individually resuspended with 1.5 mL THY, serially diluted in THY, and plated to determine the number of CFU constituting the initial time point (T0). The remaining lids were closed over the petri dish bottoms and placed in the dark at room temperature for 1, 2, 4, or 7 days until being similarly resuspended and serially diluted to determine the CFU. The percent remaining viable bacteria was calculated by dividing the time point number of CFU by the T0 number of CFU. 5.2.3 Starvation S. pneumoniae D39 was scraped off a blood agar plate and split evenly into 4 pools as described above, each of which was spread onto PBS agar (1.5% agar [Fisher] in 1X PBS [Boston Bioproducts]). The plates were closed and incubated at room 138 temperature in the dark. The percent remaining viable bacteria at 6, 24, and 48 hours was calculated as described for the desiccation experiments. 5.2.4 Competition desiccation Competition desiccations were largely conducted like single-strain desiccations, with changes listed below. An S. pneumoniae starter culture containing ~107-108 CFU of a 1:1 mixture of WT (CmS) and mutant (CmR) was thawed and plated onto blood agar. This was desiccated as described in Section 5.2.1. All samples were diluted and plated onto blood agar to enumerate total CFU. The following day, these colonies were either replica plated or patched first onto blood agar containing 2 or 3 µg/mL Cm, and then onto blood agar. After overnight growth, the ratio of mutant was determined by dividing the CmR CFU by the total CFU at each time point. The competitive index (CI) was R R calculated by dividing the time point Cm percentage by the T0 Cm percentage. 5.3 Capsule assays 5.3.1 Anti-capsule Western blot As described in the desiccation experiments, an S. pneumoniae starter culture was thawed, plated on blood agar, and grown for 16 hours before the lawn was scraped off the plate, split evenly and spread onto petri dish lids. One lid (T0) was immediately resuspended with 1.5 mL THY, and a small volume (20 µL) was used to serially dilute and plate for the number of CFU, while the remaining volume was spun for 7 minutes at 1500 x g at room temperature. The supernatant was discarded, the pellet was washed with 500 µL THY, the sample was spun as before, the supernatant was removed, and the pellet was frozen at -80˚C until use. The remaining lids were closed over the petri dish 139 bottoms and desiccated in the dark under ambient conditions for 2 or 7 days, at which time they were similarly resuspended, serially diluted, plated for the number of CFU, and spun to freeze the cell pellets. The cell pellets were thawed and resuspended in 20 µL 10 mM Tris (pH 7.5), 10 µL proteinase K buffer (50 mM EDTA, 0.5% Tween 20, 0.5% Triton X-100, 50 mM Tris [pH 8]), and 2 µL proteinase K (20 mg/mL). After 30 minutes of incubation at 37˚C, each sample was mixed with 38 µL 10 mM Tris (pH 7.5) and 17.5 µL sodium dodecyl sulfate (SDS) sample buffer (50 mM Tris [pH 6.8], 12.5 mM EDTA, 2% SDS, 10% glycerol, 1% β-mercaptoethanol, 0.02% bromophenol blue) and heated for 10 minutes at 99˚C. The samples were loaded (20 µL each) onto a 10% SDS-PAGE gel, which was run for 75 minutes at 25 V. Ponceau staining was used for 30 seconds to confirm effective proteinase K treatment before being washed off with distilled water. Using the Snap i.d. protein detection system (Millipore), the membrane was blocked with 30 mL 1X NapBlock (G Biosciences) in 1X Tris-buffered saline (TBS). The membrane was incubated for 10 minutes with polyclonal rabbit anti-type 2 capsule antiserum (Statens Serum Institut) at 1:1,000 in 1X TBS. After being washed with 90 mL 1X TBS, the membrane was incubated for 10 minutes with Cy5 goat anti-rabbit IgG (Invitrogen) at 1.7:1,000 in 1X TBS while protected from light. The membrane was washed with 90 mL 1X TBS and visualized using FLA-9000 Starion (Fujifilm). 5.3.2 Quellung A bacterial colony was transferred from the blood agar plate to a microscope slide and mixed with two 5 µL drops of dH2O. A 5 µL drop of anti-type 2 or type 4 capsule antiserum was added to one of the drops and mixed well. The cells were observed under 140 the microscope to look for swelling in the sample with antiserum added. Swelling is a positive reaction, indicating the pneumococci tested are of that capsular type. 5.4 Mouse model of nasopharyngeal colonization In all animal experiments, 8- to 12-week-old female outbred Swiss-Webster mice (Charles River Laboratories or Taconic Laboratories) were used. Mice were mock infected with 1X PBS or infected with either non-desiccated or 48-hour desiccated S. pneumoniae D39. The non-desiccated S. pneumoniae was prepared by plating one starter culture on blood agar. After 16 hours of growth, the colonies were resuspended in 1X PBS and adjusted to ~105 CFU/mL. The desiccated S. pneumoniae was prepared as described in Section 5.2.1, except at 48 hours post-desiccation, the bacteria were 5 resuspended in 1X PBS and adjusted to ~10 CFU/mL by OD600 measurement. Mice were lightly anesthetized by inhalation of 2.5% isofluorane and inoculated (5 µL/nare) with either 1X PBS, PBS-suspended non-desiccated S. pneumoniae, or PBS- suspended desiccated S. pneumoniae. Mice were killed by CO2 asphyxiation 3 days later. Bacteria colonizing the nasopharynx were recovered by lavage with 0.5 mL sterile 1X PBS through an opening made in the trachea. The lavage fluid was serially diluted and plated on blood agar containing 3 µg/mL Gent. 5.5 Library construction and Tn-seq 5.5.1 Preparation of genomic DNA Approximately 1.5 mL mid-exponential phase (OD600 0.4-0.6) S. pneumoniae was centrifuged at 4,000 rpm for 5 minutes. The supernatant (SN) was removed, and the 141 pellet resuspended in 360 µL Buffer ATL (Qiagen), after which 20 µL proteinase K was added. The sample was incubated at 56˚C for 1 hour, then 400 µL Buffer AL (Qiagen) was mixed in before transfer to an O-ring tube (Fisher) filled with ~0.5 mL 0.1mm Zirconia beads (BioSpec). The sample was bead beat for 2.5 minutes, then centrifuged at 13,200 rpm for 1.5 minutes. Carefully avoiding beads, the liquid was transferred to a new 1.5 mL microfuge tube. To this, 400 µL 100% ethanol (EtOH) and 20 µL RNAse (10 mg/mL) were added and mixed by inversion. The sample was incubated at room temperature for 10 minutes, then transferred to a DNeasy mini spin column in a 2 mL collection tube (Qiagen). The following steps are taken from the Qiagen DNeasy Tissue DNA Purification protocol. The spin column was centrifuged at 8,000 rpm for 1 minute, and the flow-through and collection tube were discarded. The spin column was transferred to a new collection tube, 500 µL Buffer AW1 (Qiagen) was added, the sample was spun as before, and the flow-through and collection tube were discarded. The spin column was transferred to a new collection tube, 500 µL Buffer AW2 (Qiagen) was added, the sample was spun at 13,200 rpm for 3 minutes, and the flow-through was discarded. The spin column was spun at 13,200 rpm for 1 minute to dry the membrane, and the collection tube was discarded. The spin column was transferred to a new 1.5 mL microfuge tube, 100-200 µL Buffer AE (Qiagen) was incubated on the membrane for 1 minute, then was spun at 8,000 rpm for 1 minute. The DNA yield was measured on the Nanodrop 1000 (Thermo). 5.5.2 In vitro transposition Genomic DNA was prepared from S. pneumoniae D39 strain AC1770, or from a pre-existing acapsular D39 (AC326) or TIGR4 (AC2394) transposon insertion library. In 142 a 1.5 mL microfuge tube, 10 µL 2X Buffer (5 µL 80% glycerol, 4 µL 20mM DTT, 4 µL BSA [New England Biolabs], and 8 µL 5X salt [100 µL 125 mM Hepes (pH 7.9), 11.5 µL 5M NaCl, and 5.75 µL 1M MgCl2]) was combined with 1 µg S. pneumoniae gDNA, 1 µg Magellan6 plasmid, and dH2O to a final volume of 19.5 µL. To this, 0.5 µL MarC9 transposase was added, and the tube was incubated at 30˚C for 1 hour, followed by heat inactivation at 75˚C for 10 minutes. The tube was incubated on ice for 5 minutes, after which 2 µL 1M NaCl and 500 µL 100% EtOH were added. The tube was inverted 10 times to thoroughly mix, then the sample was EtOH-precipitated at -20˚C for 45 minutes and -80˚C for 15 minutes. Alternatively, the tube was stored at -20˚C overnight. The tube was spun at 13,200 rpm for 30 minutes, the SN removed, and the sample was washed with 250 µL 70% EtOH. The tube was spun at 13,200 rpm for 10 minutes, the SN removed, and the sample was air-dried for approximately 30 minutes with the lid open before being dissolved in 12.5 µL dH2O. The DNA yield was measured by Nanodrop. 5.5.3 Transformation A starter culture of S. pneumoniae D39 (AC1770 or AC326) was thawed and used to seed a 5 mL culture of THY + 65 µL 1M HCl + 12.5 µL 20% glycine. After 30 to 60 minutes of growth in a 37˚C incubator with 5% CO2, 500 µL was transferred to a culture tube containing 500 µL pre-warmed THY in a 37˚C heat block. Added to the 1 mL sample was 10 µL 1M NaOH, 25 µL 8% BSA, and 1 µL 1M CaCl, with the tube briefly vortexed between each addition. Finally, 1.6 µL of competence-stimulating peptide 1 (CSP1, 50 ng/µL) was added and the tube transferred to the 37˚C CO2 incubator. After precisely 15 minutes, gDNA or PCR product (~80-300 ng) was added, then the tube was 143 briefly vortexed to mix, and placed back in the 37˚C CO2 incubator for 45 minutes of outgrowth. The sample was plated on blood agar containing the appropriate drug based on the marker located within the gDNA or PCR product (2 µg/mL Cm, 100 µg/mL Strep, or 200 µg/mL Spec). 5.5.4 Making aggregate transposon insertion library Transformant colonies from each experiment were combined into an individual small library (Table 4-1). Each blood agar plate containing transformants was washed with 1-2 mL THY, and this bacterial resuspension was transferred to a culture tube. Based on the total volume, 0.5% Oxy was added, along with 200 µg/mL Spec because R the transposon confers Spec . This was incubated for 1.5 hours at 37˚C with 5% CO2, then was transferred to a 15 mL conical tube to be spun at 4,000 rpm for 7 minutes. The SN was removed, and the pellet resuspended in 3-9 mL (higher volume for higher number of transformants) THY + 12% glycerol. This was aliquoted in 400 µL volumes in 1.5 mL microfuge tubes, and these were frozen at -80˚C. Based on the transforming genetic material, we estimated the complexity, or unique transformation events, of our recovered transformants (100% unique insertion site complexity for transformants made using in vitro transposition products, versus 50% complexity for transformants made using gDNA from an existing library). Based on the complexity of each small library, we combined different volumes of their starter cultures so that each transformant was equally represented in the aggregate library. The small library with highest complexity (Table 3-1, experiment E) was assigned a value of 1 (i.e. 100%). All other small libraries were assigned a value based on their complexity as compared with library E. For example, library H5 was given 0.33 144 because it was calculated to have 33% of the complexity (i.e. the number of unique transformants) of library E. These values were then used to determine the volumes of small library starter cultures to combine together. Library E was 600 µL, library H5 was 198 µL, etc. From all libraries, this totaled approximately 3.125 mL, which was combined THY + Oxyrase + Spec for a final volume of 50 mL. This culture was outgrown at 37˚C and 5% CO2 for 3.5 hours to mid-exponential phase (OD600 0.5), at which point it was centrifuged at 4,000 rpm for 7 minutes. The SN was removed, the pellet resuspended in 50 mL THY + 12% glycerol, and 0.5 mL aliquots were frozen at - 80˚C. These constitute the aggregate library starter cultures. 5.5.5 Two- and seven-day desiccation of aggregate library As in a desiccation, two starter cultures of the aggregate library were thawed and plated onto large blood agar plates (154 cm2) containing 200 µg/mL Spec. After 16 hours of overnight growth, the lawn was scraped off of the plate and evenly divided into three pools, which were spread onto large (165 cm2) polystyrene petri dish lids (Fisher) using a plastic straight-edge. The bacteria from one lid were immediately resuspended in 2 mL THY for T0, and a small volume was used to dilute and plate for CFU. The volume of resuspended bacteria were then split in half – half was centrifuged, washed in PBS, re- centrifuged and the SN removed prior to freezing the pellet (total cells) at -80˚C. The other half was plated onto blood agar with 200 µg/mL Spec for overnight growth, the colonies were collected by washing the plate with 2 mL THY, centrifuged, washed in PBS, and the SN removed prior to freezing the pellet (live cells). The remaining two petri dish lids were desiccated in the dark under ambient laboratory conditions for 2 (T2) or 7 (T7) days, at which point they were resuspended and treated as described for the T0 145 time point. From the frozen bacterial pellets, gDNA was made for use in Tn-seq sample preparation. 5.5.6 Four-day desiccation of aggregate library Twelve starter cultures of the aggregate library were thawed and split in half, giving 24 replicate pools of 250 µL each, which were plated onto large blood agar plates containing 200 µg/mL Spec. After 16 hours of overnight growth, 4 of the lawns (T0) were scraped up and transferred into individual 15 mL conical tubes with 2.5 mL THY each. They were thoroughly vortexed to break up bacterial clumps, and 20 µL was taken for serial dilutions and plating. To increase the starting number of bacteria from which to make gDNA, samples were pooled together in pairs, giving a final number of 2 pooled T0 samples. These were then mixed well and aliquoted into 500 µL volumes in 1.5 mL microfuge tubes, and spun at 7,000 rpm for 5 minutes. The SN was removed and the pellets were frozen at -80˚C. The remaining 20 lawns were scraped up and spread individually onto large petri dish lids, then desiccated for 4 days in the dark under ambient laboratory conditions. Each lid of bacteria was resuspended with 3 mL THY, 30 µL was taken for serial dilutions and plating, and the remaining volume was plated onto a large blood agar plate containing 200 µg/mL Spec to select for live cells. Colonies on 10 of the blood agar plates were washed with 3 mL THY, and all of them were pooled into one 50-mL conical tube. This pool was mixed thoroughly and aliquoted into 500 µL volumes in 1.5 mL microfuge tubes, spun at 7,000 rpm for 5 minutes, and pellets were frozen at -80˚C. This process was repeated for the other 10 blood agar plates, giving 2 pooled T4 samples. The frozen bacterial pellets were used to make gDNA for Tn-seq sample preparation. 146 5.5.7 MmeI method of sample preparation and Illumina sequencing Sample preparation was generally conducted as described previously (230, 231) (TvO 2009, 2010). Live-cell gDNA from the desiccation input (T0) and output (T2, T7) was digested with MmeI (NEB) in a total volume of 200 µL for 2.5 hours at 37˚C. To prevent re-ligation, the 5’ phosphate group was removed via a 1-hour incubation at 37˚C with 2 µL of calf intestinal alkaline phosphatase (NEB). The DNA was extracted with phenol:chloroform:isoamyl alcohol, precipitated with 95% alcohol, washed with 70% ethanol, and air-dried prior to being dissolved in 25 µL DEPC-treated H2O (Ambion). Using T4 DNA ligase, an adapter containing a sequencing barcode was ligated to the 2- nucleotide overhang created by MmeI in an overnight reaction at 16˚C. PCR inhibitors present in the samples were removed by Performa DTR Gel Filtration Cartridge (Edge Biosystems) according to the manufacturer’s instructions. To amplify the product, a 19- cycle PCR was performed using primers – one complementary to the adapter and the other to the inverted repeat sequence found in Magellan6 – and the adapter-ligated samples as the template. Both primers also contain Illumina-recognizable sequences. The PCR product was gel purified and eluted in 50µL 0.3X Buffer EB (Qiagen) before being run through a Performa spin column. The samples were concentrated by speed- vacuum, and the DNA yield was assessed using a high sensitivity chip on a Bioanalyzer (Agilent). The samples were mixed at 6 nM and sequenced on an Illumina Genome Analyzer II according to the manufacturer’s protocol (Illumina). Unique barcode sequences within the adapter allowed for the mixing of multiple samples in one flow cell lane and then for sample separation after sequencing. After 30 sequencing cycles, raw data was separated based on the 6-nucleotide barcode sequence, and the barcode and 4 bp 147 of the adapter sequence were removed from the data, resulting in 20-nucleotide S. pneumoniae sequences. 5.5.8 HTML method of sample preparation Sample preparation was largely conducted as described previously (121). Fifty microliters of live-cell gDNA (from before or after desiccation) was filtered through a Performa DTR Gel Filtration Cartridge. Keeping the samples on ice, the gDNA concentration was determined by Nanodrop, and 2 µg of gDNA was brought to a final volume of 20 µL with DEPC-treated water in a 2 mL microfuge tube. The tube was placed in a pre-chilled (4˚C) high-intensity cuphorn sonifier (Branson) for 2 minutes at 50% intensity using a 5 seconds on and 5 seconds off duty cycle. A 5-µL volume of the sample was run on a 2% agarose gel to confirm that the gDNA was sheared to a 100-600 bp size range. Poly-deoxycytosine (dC) tails were added to the 3’ ends of the gDNA (14.5 µL) using 0.5 µL Terminal deoxynucleotidyl Transferase (TdT, Promega), 1 µL 9.5 mM dCTP/0.5 mM ddCTP mixture, and 4 µL 5X TdT reaction buffer (Promega) in a reaction at 37˚C for 1 hour. This was heat-inactivated at 75˚C for 20 minutes, and small molecules were again removed using the Performa DTR Gel Filtration Cartridge. C-tail complementary primer olj376 and a transposon-specific primer were ligated to the gDNA by combining 5 µL gDNA, 1 µL of each primer (30 µM), 0.8 µL 25 mM dNTPs, 1 µL Easy A Cloning Enzyme (Invitrogen), 5 µL 10X Easy A Buffer (Invitrogen), and 36.2 µL DEPC-treated water under amplification conditions: 95˚C for 2 minutes; 25 cycles of 95˚C for 30 seconds, 60˚C for 30 seconds, 72˚ for 2 minutes; final extension of 72˚C for 2 minutes; hold at 4˚C. A second amplification combined 1 µL of reaction 1 product, 1 µL of 30 µM nested primer (transposon-specific), 1 µL of 30 µM index primer (contains 148 Illumina-specific barcode), 0.8 µL 25 mM dNTPs, 1 µL Easy A Cloning Enzyme, 5 µL 10X Easy A Buffer, and 40.2 µL DEPC-treated water under reaction conditions: 95˚C for 2 minutes; 15 cycles of 95˚C for 30 seconds, 60˚C for 30 seconds, 72˚C for 2 minutes; final extension of 72˚C for 2 minutes; hold at 4˚C. PCR products were put through the Qiagen PCR Clean-Up Kit per manufacturer’s instructions, and eluted in 50 µL DEPC- treated water. Five microliters was run on a 2% agarose gel to ensure that the gDNA smear was primarily between 120-600 bp in size and there were no significant primer dimers. Equivalent amounts of gDNA were multiplexed on the HiSeq2000 (Illumina). 5.5.9 Data analysis and fitness calculations For the 2-day and 7-day data, calculations were generally conducted as described previously (230). Due to a high bottleneck we attribute to loss during desiccation, we combined the two biological replicates into one sample group for analysis. The 20-bp S. pneumoniae reads resulting from Illumina sequencing and barcode removal were mapped to the S. pneumoniae D39 genome using the program Bowtie. Parameters were set allowing for up to a 1-bp mismatch per read as long as reads mapped to a unique genomic location. Any reads mapping to multiple sites in the genome were eliminated. Also, any unique insertion locations with less than 3 reads in the T0 sample were eliminated from analysis. The data were normalized by equalizing the total number of sequenced reads per time point. The change over time (T0 input compared to T2 or T7 output) in the number of reads at a specific genomic location was used to calculate fitness. For the 4-day data, calculations were based on the previously published protocol (121). Samples were de-multiplexed based on Illumina barcode, and homopolymer tails were removed from the 3’ ends of reads. The resulting reads were mapped to the S. 149 pneumoniae genome using Bowtie, allowing up to 2 mismatches in reads of at least 18 bp in length. Unmapped Bowtie reads were removed, a 15-read cutoff was imposed on the T0 data, and insertions in the last 5% of genes were excluded. The two biological replicates each of T0 and T4 were combined to allow for more robust numbers. For each insertion, fitness was calculated by comparing the fold contraction of the transposon insertion mutant relative the rest of the population (230). In other words, the data were normalized against a set of neutral genes – those that thus far have been found to be inactive in S. pneumoniae D39 and therefore have no fitness effect. To do so, the fitness over all the neutral genes was calculated, and the average was set to a value of 1. The factor used to set the neutral genes to a fitness of 1 was applied to the rest of the genome so that all fitness values are relative to the neutral value of 1. This contraction factor represents the loss of the bacterial population during the library selection, and was measured by percent bacterial survival at the time points. The combined 2-day bottleneck was 0.0367, and the combined 7-day bottleneck was 0.8571. The 4-day bottlenecks were 0.3234 and 0.3160 for the biological replicates, which were combined after normalization. All the insertions in each gene were used to calculate the overall fitness of the gene. To control for fitness deviations due to transposon insertions with a low number of reads, a weighted average was used so that insertions with under 50 reads received a proportionally lower consideration in the calculation. Three criteria were used to determine whether or not fitness values differed significantly from the WT (neutral) fitness of 1: A) fitness was calculated from at least 4 transposon insertions per gene (genes with under 4 were excluded from analysis); B) fitness deviated from WT (fitness = 1) by at least 20%; and C) fitness was significantly 150 different from WT in a one sample t test with Bonferroni correction. Of note is that after analysis, we determined that the actual number of transposon insertion mutants in our library is approximately 28,000 and not the 30,000 we had estimated. 5.6 Stress assays 5.6.1 UV-induced DNA damage A starter culture of S. pneumoniae was thawed and used to seed a 3 mL THY + Oxy culture, which was grown to mid-exponential phase (OD600 0.6) at 37˚C with 5% CO2. This was split into 500 µL aliquots in 1.5 mL microfuge tubes, and centrifuged at 10,000 rpm for 5 minutes. The SN was removed, the pellet washed in 500 µL 1X PBS, re-spun, and finally resuspended in 500 µL 1X PBS. Twenty microliters was taken for serial dilutions to establish the starting concentration, and a 50 µL droplet was placed onto a 1 in2 Parafilm square, which was put into the UV Stratalinker 1800 (Stratagene) and subjected to a controlled dose (Joules) of UV-C (254 nm) light. After UV exposure, the droplet was serially diluted and plated to determine surviving CFU. Percent survival was determined by dividing the UV-exposed CFU by the starting, untreated CFU. 5.6.2 Mid-exponential-growth-phase bacteria H2O2 assay A WT D39 starter culture was thawed and used to seed a 3 mL THY + Oxy culture, which was grown to mid-exponential phase (OD600 0.6) at 37˚C with 5% CO2. This was split into 500 µL aliquots in 1.5 mL microfuge tubes, which were centrifuged at 10,000 rpm for 5 minutes. The SN was removed and the pellet resuspended in 500 µL of THY or THY + H2O2 (10 mM, 50 mM, or 1 M, Sigma-Aldrich). Ten microliters of this resuspension was immediately taken for serial dilutions to enumerate initial CFU. All 151 serial dilutions occurred in THY only. Samples were kept at room temperature, and 10 µL volumes were taken every 20-30 minutes over 3-3.5 hours to dilute and plate for CFU. Percent survival was calculated by dividing time point CFU by initial CFU. 5.6.3 Plate-grown bacteria H2O2 assay An S. pneumoniae starter culture was thawed and diluted 1:2000 such that ~106 cells were plated onto blood agar for overnight growth at 37˚C with 5% CO2. The lawn was washed with 2 mL PBS and the bacteria were back-diluted to a mid-exponential- phase-equivalent resuspension (OD600 0.6). This was split into 500 µL aliquots in 1.5 mL microfuge tubes, which were centrifuged at 10,000 rpm for 5 minutes. The SN was removed and the pellet resuspended in 500 µL PBS +/– 10 mM H2O2 (CVS). Immediately after resuspension, 10 µL was taken for serial dilutions to determine starting CFU. All serial dilutions occurred in PBS only. Samples were kept at room temperature, and 10 µL volumes were taken every 30 minutes over 3 hours to dilute and plate for CFU. Percent survival was calculated by dividing time point CFU by initial CFU. 5.6.4 Growth curve Starter cultures of WT D39 and ∆spxB::cat were thawed and used individually to seed 5 mL cultures of THY + Oxyrase. These were OD600-matched at 0.05 using an Ultrospec 10 Classic spectrophotometer (GE Healthcare), and then each was split into 3 cultures. One received no Oxyrase, one received 0.5% Oxyrase, and one received 1% Oxyrase. They were aliquoted into 200 µL volumes in individual wells of a 96-well flat- bottom plate (n = 6 per sample group), which was placed into a plate-reader (BioTek Synergy HT) for 16 hours of growth with shaking at 37˚C. Readings were taken every 15 minutes using KC4 software. 152 5.6.5 Osmotic shock In the downshock assay, an S. pneumoniae starter culture was thawed and used to seed a 3 mL THY + Oxy culture, which was grown to mid-exponential phase (OD600 0.6) at 37˚C with 5% CO2. This was split into two 500 µL aliquots in 1.5 mL microfuge tubes, and centrifuged at 10,000 rpm for 5 minutes. The SN was removed, and one pellet was resuspended in 500 µL THY (control) and the other resuspended in 500 µL filtered dH2O (Millipore) for downshock. Immediately after resuspension, 10 µL was taken for serial dilutions to determine starting CFU. Serial dilutions occurred in the sample’s corresponding suspension medium (THY or dH2O). Samples were kept at room temperature, and 10 µL volumes were taken for diluting and plating every hour over 4 hours. Control percent survival was calculated by dividing THY-suspended time point CFU by THY initial CFU. Experimental percent survival was calculated by dividing dH2O-suspended time point CFU by THY initial CFU. In the upshock assay, a culture was similarly seeded, grown, and aliquoted. The samples were resuspended in 500 µL dH2O and incubated for 30 minutes at room temperature to equilibrate the bacteria to a low osmolar solution. They were then centrifuged at 10,000 rpm for 5 minutes and the SN was removed. One tube was resuspended in dH2O (control) and the other in Tris + 20% sucrose for osmotic upshock. Immediately after resuspension, 10 µL was taken for serial dilutions to determine starting CFU. Serial dilutions occurred in the sample’s corresponding suspension medium (dH2O or Tris + 20% sucrose). Samples were kept at room temperature, and 10 µL volumes were taken for diluting and plating after 2 and 3 hours in the new resuspension medium. 153 Percent survival was calculated by dividing the time point CFU over the corresponding starting CFU. 5.7 Statistical analysis Survival data of 48-hour desiccated acapsular (AC326) and encapsulated (AC1770) strain D39 were shown to be normally distributed by the D’Agostino-Pearson omnibus normality test, and differences were then tested using the unpaired Student t test. The Mann-Whitney U test was used to test differences in desiccation survival of mid- exponential-phase versus plate-grown bacteria, differences in colonization of mice, and differences in single-strain desiccation survival. Wilcoxon Signed Rank test was used to test the difference of the CI from neutral value of 1. One-way analysis of variance (ANOVA) (Kruskal-Wallis test with Dunn’s post-test) was used to test differences in survival between starved and desiccated bacteria, and to interpret data from the UV, plate-grown H2O2, and osmotic downshock assays. Outliers were assessed by Grubb’s test for a single outlier (contchart.com/outliers.aspx) and removed from data sets and subsequent statistical tests if p < 0.05. 154 Serotype Reference Strain Genotype or Description (type of variant) or source AC326 Acapsular D39 2 (207) AC353 TIGR4, StrepR derivative 4 (77) AC1356 Clinical isolate 1 J. Weiser AC1357 Clinical isolate 6A (opaque) J. Weiser AC1358 Clinical isolate 6A (transparent) J. Weiser AC1359 Clinical isolate 6B (opaque) J. Weiser AC1360 Clinical isolate 6B (transparent) J. Weiser AC1361 Clinical isolate 5 J. Weiser AC1362 Clinical isolate 23F J. Weiser AC1363 Clinical isolate 8 J. Weiser AC1364 Clinical isolate 14 J. Weiser AC1365 Clinical isolate 9V J. Weiser AC1366 Clinical isolate 18C (opaque) J. Weiser AC1367 Clinical isolate 18C (transparent) J. Weiser AC1368 Clinical isolate 3 J. Weiser AC1369 Clinical isolate 19F J. Weiser AC1371 Clinical isolate 15 J. Weiser AC1770 D39 2 (234) AC2394 Acapsular AC353, StrepR derivative 4 R. Iyer AC4974 D39 ∆mfd::cat, CmR 2 This work AC4975 D39 ∆uvrA::cat, CmR 2 This work AC4976 D39 ∆spd_00185::cat, CmR 2 This work AC4977 D39 ∆spd_00475::cat, CmR 2 This work AC4978 D39 ∆uvrC::cat, CmR 2 This work AC4979 D39 ∆spd_0563::cat, CmR 2 This work AC4980 D39 ∆spxB::cat, CmR 2 This work AC4981 D39 ∆sodA::cat, CmR 2 This work AC4982 D39 ∆mscL::cat, CmR 2 This work AC4983 D39 ∆uvrB::cat, CmR 2 This work AC4984 D39 ∆spd_1168/9::cat, CmR 2 This work AC4985 D39 ∆mscS::cat, CmR 2 This work AC4986 D39 ∆spd_1836::cat, CmR 2 This work AC4988 D39 ∆spxB::cat ∆spd_0563::spxB, CmR SpecR 2 This work AC4989 D39 ∆sodA::cat ∆spd_0563::sodA, CmR SpecR 2 This work AC4990 D39 ∆uvrB::cat ∆spd_0563::uvrB, CmR SpecR 2 This work AC4991 D39 ∆mscS::cat ∆spd_0563::mscS, CmR SpecR 2 This work RW1 D39 ∆spd_0563::spec, SpecR 2 This work RW16 D39 ∆spd_1135::cat, CmR 2 This work Table 5-1: Strains used in this study. 155 Strain Primer Name Sequence (5’ to 3’) AC4974 spd_0006 F1 TAGTCGTTATTTCTGCGCG spd_0006 R1 ATACCGTCGACCTCGAGATCTTATCCGTTATACCTCTGCATTG spd_0006 F2 CTAATGACTGGCTTTTATAATATTTTTCTTCTATAAAATAGATAAAAATGG spd_0006 R2 GCATAAACAGGAACCTCTCC AC4975 spd_0176 F1 ATTATCCTTAAAGTTCATCCCG spd_0176 R1 ATACCGTCGACCTCGAGATCCTTTGTTCTTTCTAGTCCATTATTG spd_0176 F2 CTAATGACTGGCTTTTATAAATGAATAAACGCGTACAAGC spd_0176 R2 GTAGTCACCATAACCTGCC AC4976 spd_0185 F1 TTTATCAAAGAAAAAGAGCAGG spd_0185 R1 ATACCGTCGACCTCGAGATCATCCCCTCCTTGATAGTTTG spd_0185 F2 CTAATGACTGGCTTTTATAAGAAATCTAGAACTTAGGATACAAACC spd_0185 R2 ACAAGTGAAAAACCTAAAGCC AC4977 spd_0475 F1 GATTCTGGTTAAGACAGGCC spd_0475 R1 ATACCGTCGACCTCGAGATCAAAATTATCCTCCTCAACAAACAG spd_0475 F2 CTAATGACTGGCTTTTATAAAAATTATGAAAAGGTAAGCAGAAGACG spd_0475 R2 ACGCAAACCATGAATCTCCC AC4978 spd_0538 F1 CTTTCCGTTACCATCCATCTCC spd_0538 R1 TTGTCTAAATCAATTTTATTAAAGTTCATAGTTCCATTATAGCAAAAAAAGTAGGG spd_0538 F2 CTAATGACTGGCTTTTATAAGGAAGGAAACCACAATGAACC spd_0538 R2 CTTCACATCCTCTACATACCC AC4979 spd_0563 F1 ATACTTCTGAACATCCGGCC spd_0563 R1 ATACCGTCGACCTCGAGATCGGGATAGCCTTTCTAAATGG (cat) spd_0563 F2 CTAATGACTGGCTTTTATAAAACCTGTCTTTGGGCAGATAAAGG (cat) spd_0563 R2 ACCTATAAAGGTCAACGGCC AC4980 spd_0636 F1 TGGAAATGCTAAAATGCTGG spd_0636 R1 ATACCGTCGACCTCGAGATCAATGATAACTCTCCTTCAATTTTTTTAA spd_0636 F2 CTAATGACTGGCTTTTATAATTCCTCTCGCCGAAAATC spd_0636 R2 GTGGATGTAATACTGAGTGGA AC4981 spd_0667 F1 ACAATTCTTCTCCAACAACC spd_0667 R1 ATACCGTCGACCTCGAGATCCTGTAATACCTCTTTTTCTTTCTATATG spd_0667 F2 CTAATGACTGGCTTTTATAATGATAGTTGGAGGGAAGAATTG spd_0667 R2 TTCTTCAAATGACTGCACAC AC4982 spd_0896 F1 AAATGATCTTTTCGAATCAATCCG spd_0896 R1 ATACCGTCGACCTCGAGATCACTGTTTTCCTTTCAATGTTTCG spd_0896 F2 CTAATGACTGGCTTTTATAATCGATTAAAAGGATCTAGCGC spd_0896 R2 CCTCTGCAAAAAATTACCCG AC4983 spd_1096 F1 AGTTTAAACATTTACACAAGCCC spd_1096 R1 ATACCGTCGACCTCGAGATCAAGAATCCTCTCTCCATGTGC spd_1096 F2 CTAATGACTGGCTTTTATAAGGGAATAGTATGATTTATTTAAGAAAG spd_1096 R2 TGTCCATCAAAAGTAAAACCC AC4984 spd_1168&1169 CTTCTATTCAACAAACAAACCTC F1 spd_1168&1169 ATACCGTCGACCTCGAGATCTTTAATTTCTCCTGAAAAAACAAGAG R1 (Pcat) spd_1168&1169 CTAATGACTGGCTTTTATAAGAGGTTGACAAAATGAATCAAAC F2 spd_1168&1169 CTTACAAAATGTCCGTCACC R2 156 AC4985 spd_1562 F1 TGATAGTAGAACCGAGTGGG spd_1562 R1 ATACCGTCGACCTCGAGATCCAGTCTCATTCTCCATAAATATTC spd_1562 F2 CTAATGACTGGCTTTTATAAACCTGTTATTTTAAGTAATTTGTGGTAC spd_1562 R2 GCTCCTTTTCTTAACTTTTTTTCGC AC4986 spd_1836 F1 AAATATGTTGGTCTCGATGG spd_1836 R1 ATACCGTCGACCTCGAGATCCTTAACTCTCCTTTTCTAAACGTTC spd_1836 F2 CTAATGACTGGCTTTTATAAATGAAGGAACTTTTAAACAAGGC spd_1836 R2 TGATTTTTCAACATAAGCTGG AC4988 F_spec CGCTCTAGAACTAGTGGATCCC R_spec TTATAATTTTTTTAATCTGTTATTTA spd_0563 F1 ATACTTCTGAACATCCGGCC R0563 spec tail CTATTTAAATAACAGATTAAAAAAATTATAAGGGATAGCCTTTCTAAATGG Rspec 0563 tail CCATTTAGAAAGGCTATCCCTTATAATTTTTTTAATCTGTTATTTAAATAG F0636 spec tail GGGATCCACTAGTTCTAGAGCGAAGTAATGACTAGATTTCTTTGTTATAAAAC R0636 no tail TTATTTAATTGCGCGTGATTGC R0636 CAATCACGCGCAATTAAATAAAACCTGTCTTTGGGCAGATAAAGG spd_0563 R2 ACCTATAAAGGTCAACGGCC AC4989 F_spec CGCTCTAGAACTAGTGGATCCC R_spec TTATAATTTTTTTAATCTGTTATTTA spd_0563 F1 ATACTTCTGAACATCCGGCC R0563 spec tail CTATTTAAATAACAGATTAAAAAAATTATAAGGGATAGCCTTTCTAAATGG Rspec 0563 tail CCATTTAGAAAGGCTATCCCTTATAATTTTTTTAATCTGTTATTTAAATAG F0667 spec tail GGGATCCACTAGTTCTAGAGCGTCAGTGATTTCCTTATCTATAAGC R0667 0563 tail CCTTTATCTGCCCAAAGACAGGTTTTATTTAGCAGCTGCGTAC spd_0563 F2 no AACCTGTCTTTGGGCAGATAAAGG tail spd_0563 R2 ACCTATAAAGGTCAACGGCC AC4990 F_spec CGCTCTAGAACTAGTGGATCCC R_spec TTATAATTTTTTTAATCTGTTATTTA spd_0563 F1 ATACTTCTGAACATCCGGCC R0563 spec tail CTATTTAAATAACAGATTAAAAAAATTATAAGGGATAGCCTTTCTAAATGG Rspec 0563 tail CCATTTAGAAAGGCTATCCCTTATAATTTTTTTAATCTGTTATTTAAATAG F1096 spec tail GGGATCCACTAGTTCTAGAGCGCTTCGGTCCGTCCTTTTC R1096 0563 tail CCTTTATCTGCCCAAAGACAGGTTCTAATCCAAGGCCTTGACTTC spd_0563 F2 no AACCTGTCTTTGGGCAGATAAAGG tail spd_0563 R2 ACCTATAAAGGTCAACGGCC AC4991 F_spec CGCTCTAGAACTAGTGGATCCC R_spec TTATAATTTTTTTAATCTGTTATTTA spd_0563 F1 ATACTTCTGAACATCCGGCC R0563 spec tail CTATTTAAATAACAGATTAAAAAAATTATAAGGGATAGCCTTTCTAAATGG Rspec 0563 tail CCATTTAGAAAGGCTATCCCTTATAATTTTTTTAATCTGTTATTTAAATAG F1562 spec tail GGGATCCACTAGTTCTAGAGCGTTCCAAGCATGAATGATTC R1562 0563 tail CCTTTATCTGCCCAAAGACAGGTTCTAGTCTGTGCGTGAGAAATTG spd_0563 F2 no AACCTGTCTTTGGGCAGATAAAGG tail spd_0563 R2 ACCTATAAAGGTCAACGGCC RW1 spd_0563 F1 ATACTTCTGAACATCCGGCC spd_0563 R1 GGATCCACTAGTTCTAGAGCGGGGATAGCCTTTCTAAATGG spd_0563 F2 TAAATAACAGATTAAAAAAATTATAAAACCTGTCTTTGGGCAGATAAAGG spd_0563 R2 ACCTATAAAGGTCAACGGCC 157 RW16 spd_1135 F1 TTTCTGTTTTTTGAGCTCTTGC spd_1135 R1 ATACCGTCGACCTCGAGATCTTATTCATCTCCGTCAAATAGTCC spd_1135 F2 CTAATGACTGGCTTTTATAAGATGCTACATTGAAACTTTTTGC spd_1135 R2 CTTCATAGGGAATCTGGGC Table 5-2: Primers used in this study. 158 APPENDIX: Fitness Tables 159 Table A-1: DNA repair genes. Listed are genes in DNA repair pathways (4), with their fitness from analysis of the 2-day and 4-day aggregate library desiccations. The “Insertions” columns list the number of unique transposon insertion sites in each gene used to calculate the fitness. Genes with 0 insertions do not have a calculated fitness. 160 Table A-2: Significant genes identified by Tn-seq found in D39, but not TIGR4. Listed are genes with fitnesses significantly different from the WT value of 1 (at least 10% different) after analysis of the 2-day and 4-day aggregate library desiccations. These D39 genes do not have corresponding homologues in TIGR4. Highlighted fitnesses are significant at that time point. (p value determined by Student t test with Bonferroni correction) 161 Table A-3: Significant genes identified by desiccation Tn-seq, but unresponsive in other Tn-seq conditions. Listed are genes with fitnesses significantly different from the WT value of 1 (at least 10% different) after analysis of the 2-day and 4-day aggregate library desiccations. These genes were significant in at least one of the two desiccation time points, but their homologues were unresponsive in all 20 in vitro and in vivo conditions previously tested by Tim van Opijnen (230, 232). Fitnesses highlighted in red (lower than 1) and green (higher than 1) are significant at that time point. (p value determined by Student t test with Bonferroni correction) 162 REFERENCES 1. Akinyemi, K. O., A. D. Atapu, O. O. Adetona and A. O. Coker. 2009. The potential role of mobile phones in the spread of bacterial infections. J Infect Dev Ctries 3:628-632. 2. AlonsoDeVelasco, E., A. F. Verheul, J. Verhoef and H. Snippe. 1995. Streptococcus pneumoniae: virulence factors, pathogenesis, and vaccines. Microbiol Rev 59:591-603. 3. Alothman, A., A. Jelani, A. Althaqafi, M. Rich and E. Williams. 2003. Contamination of patient hospital charts by bacteria. J Hosp Infect 55:304-305. 4. Ambur, O. H., T. Davidsen, S. A. Frye, S. V. Balasingham, K. Lagesen, T. Rognes and T. Tonjum. 2009. Genome dynamics in major bacterial pathogens. FEMS Microbiol Rev 33:453-470. 5. Anderson, D. G. and S. C. Kowalczykowski. 1997. The translocating RecBCD enzyme stimulates recombination by directing RecA protein onto ssDNA in a chi- regulated manner. Cell 90:77-86. 6. Andisi, V. F., C. A. Hinojosa, A. de Jong, O. P. Kuipers, C. J. Orihuela and J. J. Bijlsma. 2012. Pneumococcal gene complex involved in resistance to extracellular oxidative stress. Infect Immun 80:1037-1049. 7. Ansari, S. A., V. S. Springthorpe and S. A. Sattar. 1991. Survival and vehicular spread of human rotaviruses: possible relation to seasonality of outbreaks. Rev Infect Dis 13:448-461. 8. Aricha, B., I. Fishov, Z. Cohen, N. Sikron, S. Pesakhov, I. Khozin-Goldberg, R. Dagan and N. Porat. 2004. Differences in membrane fluidity and fatty acid composition between phenotypic variants of Streptococcus pneumoniae. J Bacteriol 186:4638-4644. 9. Asada, S., M. Takano and I. Shibasaki. 1979. Deoxyribonucleic acid strand breaks during drying of Escherichia coli on a hydorohobic filter membrane. Appl Environ Microbiol 37:266-273 10. Asada, S., M. Takano and I. Shibasaki. 1980. Mutation induced by drying of Escherichia coli on a hydrophobic filter membrane. Appl Environ Microbiol 40:274-281. 11. Battesti, A., N. Majdalani and S. Gottesman. 2011. The RpoS-mediated general stress response in Escherichia coli. Annu Rev Microbiol 65:189-213. 12. Bauermeister, A., R. Moeller, G. Reitz, S. Sommer and P. Rettberg. 2011. Effect of relative humidity on Deinococcus radiodurans' resistance to prolonged desiccation, heat, ionizing, germicidal, and environmentally relevant UV radiation. Microb Ecol 61:715-722. 13. Bebbington, A., I. Parkin, P. A. James, L. J. Chichester and E. M. Kubiak. 2003. Patients' case-notes: look but don't touch. J Hosp Infect 55:299-301. 14. Beblo, K., E. Rabbow, R. Rachel, H. Huber and P. Rettberg. 2009. Tolerance of thermophilic and hyperthermophilic microorganisms to desiccation. Extremophiles 13:521-531. 15. Benisty, R., A. Y. Cohen, A. Feldman, Z. Cohen and N. Porat. 2010. Endogenous H2O2 produced by Streptococcus pneumoniae controls FabF activity. Biochim Biophys Acta 1801:1098-1104. 163 16. Billi, D. and M. Potts. 2002. Life and death of dried prokaryotes. Res Microbiol 153:7-12. 17. Billi, D., D. J. Wright, R. F. Helm, T. Prickett, M. Potts and J. H. Crowe. 2000. Engineering desiccation tolerance in Escherichia coli. Appl Environ Microbiol 66:1680-1684. 18. Blythe, D., D. Keenlyside, S. J. Dawson and A. Galloway. 1998. Environmental contamination due to methicillin-resistant Staphylococcus aureus (MRSA). J Hosp Infect 38:67-69. 19. Bogaert, D., R. De Groot and P. W. Hermans. 2004. Streptococcus pneumoniae colonisation: the key to pneumococcal disease. Lancet Infect Dis 4:144-154. 20. Bortoni, M. E., V. S. Terra, J. Hinds, P. W. Andrew and H. Yesilkaya. 2009. The pneumococcal response to oxidative stress includes a role for Rgg. Microbiology 155:4123-4134. 21. Brown, E. J., K. A. Joiner, R. M. Cole and M. Berger. 1983. Localization of complement component 3 on Streptococcus pneumoniae: anti-capsular antibody causes complement deposition on the pneumococcal capsule. Infect Immun 39:403-409. 22. Bures, S., J. T. Fishbain, C. F. Uyehara, J. M. Parker and B. W. Berg. 2000. Computer keyboards and faucet handles as reservoirs of nosocomial pathogens in the intensive care unit. Am J Infect Control 28:465-471. 23. Buttery, J. P., S. J. Alabaster, R. G. Heine, S. M. Scott, R. A. Crutchfield, A. Bigham, S. N. Tabrizi and S. M. Garland. 1998. Multiresistant Pseudomonas aeruginosa outbreak in a pediatric oncology ward related to bath toys. Pediatr Infect Dis J 17:509-513. 24. Carapetis, J. R., A. C. Steer, E. K. Mulholland and M. Weber. 2005. The global burden of group A streptococcal diseases. Lancet Infect Dis 5:685-694. 25. Carson, D., R. A. Pieringer and L. Daneo-Moore. 1979. Effect of growth rate on lipid and lipoteichoic acid composition in Streptococcus faecium. Biochim Biophys Acta 575:225-233. 26. Centers for Disease Control and Prevention US. 2011. Epidemiology and Prevention of Vaccine-Preventable Diseases: Pneumococcal Disease. Public Health Foundation (12th Edition). 233-248. 27. Chaibenjawong, P. and S. J. Foster. 2011. Desiccation tolerance in Staphylococcus aureus. Arch Microbiol 193:125-135. 28. Chen, K. H., L. R. Chen and Y. K. Wang. 2014. Contamination of medical charts: an important source of potential infection in hospitals. PLoS One 9:e78512. 29. Cheng, Q., E. A. Campbell, A. M. Naughton, S. Johnson and H. R. Masure. 1997. The com locus controls genetic transformation in Streptococcus pneumoniae. Mol Microbiol 23:683-692. 30. Choi, I. H., J. H. Shim, S. W. Kim, S. N. Kim, S. N. Pyo and D. K. Rhee. 1999. Limited stress response in Streptococcus pneumoniae. Microbiol Immunol 43:807-812. 31. Claverys, J. P., M. Prudhomme and B. Martin. 2006. Induction of competence regulons as a general response to stress in gram-positive bacteria. Annu Rev Microbiol 60:451-475. 164 32. Cosgrove, K., G. Coutts, I. M. Jonsson, A. Tarkowski, J. F. Kokai-Kun, J. J. Mond and S. J. Foster. 2007. Catalase (KatA) and alkyl hydroperoxide reductase (AhpC) have compensatory roles in peroxide stress resistance and are required for survival, persistence, and nasal colonization in Staphylococcus aureus. J Bacteriol 189:1025-1035. 33. Cox, C. S. 1970. Aerosol survival of Escherichia coli B disseminated from the dry state. Appl Microbiol 19:604-607. 34. Cox, C. S. 1971. Aerosol survival of Pasteurella tularensis disseminated from the wet and dry states. Appl Microbiol 21:482-486. 35. Croucher, N. J., J. A. Finkelstein, S. I. Pelton, P. K. Mitchell, G. M. Lee, J. Parkhill, S. D. Bentley, W. P. Hanage and M. Lipsitch. 2013. Population genomics of post-vaccine changes in pneumococcal epidemiology. Nat Genet 45:656-663. 36. Croucher, N. J., S. R. Harris, C. Fraser, M. A. Quail, J. Burton, M. van der Linden, L. McGee, A. von Gottberg, J. H. Song, K. S. Ko, B. Pichon, S. Baker, C. M. Parry, L. M. Lambertsen, D. Shahinas, D. R. Pillai, T. J. Mitchell, G. Dougan, A. Tomasz, K. P. Klugman, J. Parkhill, W. P. Hanage and S. D. Bentley. 2011. Rapid pneumococcal evolution in response to clinical interventions. Science 331:430-434. 37. Crowe, L. M. and J. H. Crowe. 1992. Anhydrobiosis: a strategy for survival. Adv Space Res 12:239-247. 38. Cruickshank, C. C., R. F. Minchin, A. C. Le Dain and B. Martinac. 1997. Estimation of the pore size of the large-conductance mechanosensitive ion channel of Escherichia coli. Biophys J 73:1925-1931. 39. Dagan, R. and K. P. Klugman. 2008. Impact of conjugate pneumococcal vaccines on antibiotic resistance. Lancet Infect Dis 8:785-795. 40. Dagkessamanskaia, A., M. Moscoso, V. Henard, S. Guiral, K. Overweg, M. Reuter, B. Martin, J. Wells and J. P. Claverys. 2004. Interconnection of competence, stress and CiaR regulons in Streptococcus pneumoniae: competence triggers stationary phase autolysis of ciaR mutant cells. Mol Microbiol 51:1071- 1086. 41. Daly, M. J., E. K. Gaidamakova, V. Y. Matrosova, A. Vasilenko, M. Zhai, R. D. Leapman, B. Lai, B. Ravel, S. M. Li, K. M. Kemner and J. K. Fredrickson. 2007. Protein oxidation implicated as the primary determinant of bacterial radioresistance. PLoS Biol 5:e92. 42. Denich, T. J., L. A. Beaudette, H. Lee and J. T. Trevors. 2003. Effect of selected environmental and physico-chemical factors on bacterial cytoplasmic membranes. J Microbiol Methods 52:149-182. 43. Dose, K., A. Bieger-Dose, M. Labusch and M. Gill. 1992. Survival in extreme dryness and DNA-single-strand breaks. Adv Space Res 12:221-229. 44. Dowell, S. F., C. G. Whitney, C. Wright, C. E. Rose, Jr. and A. Schuchat. 2003. Seasonal patterns of invasive pneumococcal disease. Emerg Infect Dis 9:573-579. 45. Dunklin, E. W. and T. T. Puck. 1948. The lethal effect of relative humidity on air-borne bacteria. J Exp Med 87:87-101. 165 46. Edwards, M. D., W. Bartlett and I. R. Booth. 2008. Pore mutations of the Escherichia coli MscS channel affect desensitization but not ionic preference. Biophys J 94:3003-3013. 47. Eiref, S. D., I. M. Leitman and W. Riley. 2012. Hand sanitizer dispensers and associated hospital-acquired infections: friend or fomite? Surg Infect (Larchmt) 13:137-140. 48. Engelmoer, D. J. and D. E. Rozen. 2011. Competence increases survival during stress in Streptococcus pneumoniae. Evolution 65:3475-3485. 49. Ermakov, Y. A., K. Kamaraju, K. Sengupta and S. Sukharev. 2010. Gadolinium ions block mechanosensitive channels by altering the packing and lateral pressure of anionic lipids. Biophys J 98:1018-1027. 50. Feldman, C. and R. Anderson. 2014. Recent advances in our understanding of Streptococcus pneumoniae infection. F1000Prime Rep 6:82. 51. Finn, S., K. Handler, O. Condell, A. Colgan, S. Cooney, P. McClure, A. Amezquita, J. C. Hinton and S. Fanning. 2013. ProP is required for the survival of desiccated Salmonella enterica serovar Typhimurium cells on a stainless steel surface. Appl Environ Microbiol 79:4376-4384. 52. Fredrickson, J. K., S. M. Li, E. K. Gaidamakova, V. Y. Matrosova, M. Zhai, H. M. Sulloway, J. C. Scholten, M. G. Brown, D. L. Balkwill and M. J. Daly. 2008. Protein oxidation: key to bacterial desiccation resistance? Isme J 2:393-403. 53. Frees, D., S. N. Qazi, P. J. Hill and H. Ingmer. 2003. Alternative roles of ClpX and ClpP in Staphylococcus aureus stress tolerance and virulence. Mol Microbiol 48:1565-1578. 54. Garcia, A. H. 2011. Anhydrobiosis in bacteria: from physiology to applications. J Biosci 36:939-950. 55. Garg, S. K., S. Kommineni, L. Henslee, Y. Zhang and P. Zuber. 2009. The YjbH protein of Bacillus subtilis enhances ClpXP-catalyzed proteolysis of Spx. J Bacteriol 191:1268-1277. 56. Garmiri, P., K. E. Coles, T. J. Humphrey and T. A. Cogan. 2008. Role of outer membrane lipopolysaccharides in the protection of Salmonella enterica serovar Typhimurium from desiccation damage. FEMS Microbiol Lett 281:155- 159. 57. Gasc, A. M., N. Sicard, J. P. Claverys and A. M. Sicard. 1980. Lack of SOS repair in Streptococcus pneumoniae. Mutat Res 70:157-165. 58. Gayoso, C. M., J. Mateos, J. A. Mendez, P. Fernandez-Puente, C. Rumbo, M. Tomas, O. Martinez de Ilarduya and G. Bou. 2014. Molecular mechanisms involved in the response to desiccation stress and persistence in Acinetobacter baumannii. J Proteome Res 13:460-476. 59. Georgiou, G. and L. Masip. 2003. Biochemistry. An overoxidation journey with a return ticket. Science 300:592-594. 60. Goldfine, H. 1972. Comparative aspects of bacterial lipids. Adv Microb Physiol 8:1-58. 61. Goldfine, H. 1982. Lipids of Prokaryotes - Structure and Distribution. In F. Bronner, S. Razin and S. Rottem (Eds.), Current Topics in Membranes and Transport: Membrane Lipids of Prokaryotes: 1-43. Academic Press, New York. 166 62. Gonzalez, M. and R. Woodgate. 2002. The "tale" of UmuD and its role in SOS mutagenesis. Bioessays 24:141-148. 63. Gray, B. M., M. E. Turner and H. C. Dillon, Jr. 1982. Epidemiologic studies of Streptococcus pneumoniae in infants. The effects of season and age on pneumococcal acquisition and carriage in the first 24 months of life. Am J Epidemiol 116:692-703. 64. Gritzfeld, J. F., A. D. Wright, A. M. Collins, S. H. Pennington, A. K. Wright, A. Kadioglu, D. M. Ferreira and S. B. Gordon. 2013. Experimental human pneumococcal carriage. J Vis Exp 65. Gruzdev, N., M. McClelland, S. Porwollik, S. Ofaim, R. Pinto and S. Saldinger-Sela. 2012. Global transcriptional analysis of dehydrated Salmonella enterica serovar Typhimurium. Appl Environ Microbiol 78:7866-7875. 66. Gruzdev, N., R. Pinto and S. Sela. 2011. Effect of desiccation on tolerance of Salmonella enterica to multiple stresses. Appl Environ Microbiol 77:1667-1673. 67. Guerzoni, M. E., R. Lanciotti and P. S. Cocconcelli. 2001. Alteration in cellular fatty acid composition as a response to salt, acid, oxidative and thermal stresses in Lactobacillus helveticus. Microbiology 147:2255-2264. 68. Guiral, S., T. J. Mitchell, B. Martin and J. P. Claverys. 2005. Competence- programmed predation of noncompetent cells in the human pathogen Streptococcus pneumoniae: genetic requirements. Proc Natl Acad Sci U S A 102:8710-8715. 69. Habimana, O., T. Moretro, S. Langsrud, L. K. Vestby, L. L. Nesse and E. Heir. 2010. Micro ecosystems from feed industry surfaces: a survival and biofilm study of Salmonella versus host resident flora strains. BMC Vet Res 6:48. 70. Hagensee, M. E. and R. E. Moses. 1989. Multiple pathways for repair of hydrogen peroxide-induced DNA damage in Escherichia coli. J Bacteriol 171:991-995. 71. Hakenbeck, R., N. Balmelle, B. Weber, C. Gardes, W. Keck and A. de Saizieu. 2001. Mosaic genes and mosaic chromosomes: intra- and interspecies genomic variation of Streptococcus pneumoniae. Infect Immun 69:2477-2486. 72. Halpern, D., A. Gruss, J. P. Claverys and M. El-Karoui. 2004. rexAB mutants in Streptococcus pneumoniae. Microbiology 150:2409-2414. 73. Halton, K., V. Arora, V. Singh, S. S. Ghantoji, D. N. Shah and K. W. Garey. 2011. Bacterial colonization on writing pens touched by healthcare professionals and hospitalized patients with and without cleaning the pen with alcohol-based hand sanitizing agent. Clin Microbiol Infect 17:868-869. 74. Hamburger, M., Jr. and M. J. Green. 1946. The problem of the dangerous carrier of hemolytic streptococci; observations upon the role of the hands, of blowing the nose, of sneezing, and of coughing in the dispersal of these microorganisms. J Infect Dis 79:33-44. 75. Hammerschmidt, S., S. Wolff, A. Hocke, S. Rosseau, E. Muller and M. Rohde. 2005. Illustration of pneumococcal polysaccharide capsule during adherence and invasion of epithelial cells. Infect Immun 73:4653-4667. 76. Hamza, N., G. Bazoua, Y. Al-Shajerie, E. Kubiak, P. James and C. Wong. 2007. A prospective study of the case-notes of MRSA-positive patients: a vehicle of MRSA spread. Ann R Coll Surg Engl 89:665-667. 167 77. Hava, D. L. and A. Camilli. 2002. Large-scale identification of serotype 4 Streptococcus pneumoniae virulence factors. Mol Microbiol 45:1389-1406. 78. Hava, D. L., J. LeMieux and A. Camilli. 2003. From nose to lung: the regulation behind Streptococcus pneumoniae virulence factors. Mol Microbiol 50:1103-1110. 79. Heber, S. and B. E. Tropp. 1991. Genetic regulation of cardiolipin synthase in Escherichia coli. Biochim Biophys Acta 1129:1-12. 80. Henriques-Normark, B., C. Blomberg, J. Dagerhamn, P. Battig and S. Normark. 2008. The rise and fall of bacterial clones: Streptococcus pneumoniae. Nat Rev Microbiol 6:827-837. 81. Hicks, L. A., L. H. Harrison, B. Flannery, J. L. Hadler, W. Schaffner, A. S. Craig, D. Jackson, A. Thomas, B. Beall, R. Lynfield, A. Reingold, M. M. Farley and C. G. Whitney. 2007. Incidence of pneumococcal disease due to non- pneumococcal conjugate vaccine (PCV7) serotypes in the United States during the era of widespread PCV7 vaccination, 1998-2004. J Infect Dis 196:1346-1354. 82. Hingston, P. A., E. C. Stea, S. Knochel and T. Hansen. 2013. Role of initial contamination levels, biofilm maturity and presence of salt and fat on desiccation survival of Listeria monocytogenes on stainless steel surfaces. Food Microbiol 36:46-56. 83. Hodges, R. G. and L. C. Mac. 1946. Epidemic pneumococcal pneumonia; the influence of population characteristics and environment. Am J Hyg 44:193-206. 84. Hoffmann, O., J. Zweigner, S. H. Smith, D. Freyer, C. Mahrhofer, E. Dagand, E. I. Tuomanen and J. R. Weber. 2006. Interplay of pneumococcal hydrogen peroxide and host-derived nitric oxide. Infect Immun 74:5058-5066. 85. Hoffmann, T., C. Boiangiu, S. Moses and E. Bremer. 2008. Responses of Bacillus subtilis to hypotonic challenges: physiological contributions of mechanosensitive channels to cellular survival. Appl Environ Microbiol 74:2454- 2460. 86. Horton, R. M., H. D. Hunt, S. N. Ho, J. K. Pullen and L. R. Pease. 1989. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77:61-68. 87. Humann, J. L., H. T. Ziemkiewicz, S. N. Yurgel and M. L. Kahn. 2009. Regulatory and DNA repair genes contribute to the desiccation resistance of Sinorhizobium meliloti Rm1021. Appl Environ Microbiol 75:446-453. 88. Ibrahim, Y. M., A. R. Kerr, J. McCluskey and T. J. Mitchell. 2004. Role of HtrA in the virulence and competence of Streptococcus pneumoniae. Infect Immun 72:3584-3591. 89. Ibrahim, Y. M., A. R. Kerr, N. A. Silva and T. J. Mitchell. 2005. Contribution of the ATP-dependent protease ClpCP to the autolysis and virulence of Streptococcus pneumoniae. Infect Immun 73:730-740. 90. Imlay, J. A. and S. Linn. 1986. Bimodal pattern of killing of DNA-repair- defective or anoxically grown Escherichia coli by hydrogen peroxide. J Bacteriol 166:519-527. 91. Imlay, J. A. and S. Linn. 1987. Mutagenesis and stress responses induced in Escherichia coli by hydrogen peroxide. J Bacteriol 169:2967-2976. 168 92. Jain, N. K. and I. Roy. 2009. Effect of trehalose on protein structure. Protein Sci 18:24-36. 93. Janion, C. 2001. Some aspects of the SOS response system--a critical survey. Acta Biochim Pol 48:599-610. 94. Janning, B. i. t. V. P. H. 1994. Susceptibility of bacterial strains to desiccation: a simple method to test their stability in microbiological reference materials. Analytica Chimica Acta 286:469-476. 95. Jawad, A., J. Heritage, A. M. Snelling, D. M. Gascoyne-Binzi and P. M. Hawkey. 1996. Influence of relative humidity and suspending menstrua on survival of Acinetobacter spp. on dry surfaces. J Clin Microbiol 34:2881-2887. 96. Jawad, A., H. Seifert, A. M. Snelling, J. Heritage and P. M. Hawkey. 1998. Survival of Acinetobacter baumannii on dry surfaces: comparison of outbreak and sporadic isolates. J Clin Microbiol 36:1938-1941. 97. Jiricny, J. 2013. Postreplicative mismatch repair. Cold Spring Harb Perspect Biol 5:a012633. 98. Johnsborg, O. and L. S. Havarstein. 2009. Regulation of natural genetic transformation and acquisition of transforming DNA in Streptococcus pneumoniae. FEMS Microbiol Rev 33:627-642. 99. Johnson, H. L., M. Deloria-Knoll, O. S. Levine, S. K. Stoszek, L. Freimanis Hance, R. Reithinger, L. R. Muenz and K. L. O'Brien. 2010. Systematic evaluation of serotypes causing invasive pneumococcal disease among children under five: the pneumococcal global serotype project. PLoS Med 7: 100. Johnston, C., N. Campo, M. J. Berge, P. Polard and J. P. Claverys. 2014. Streptococcus pneumoniae, le transformiste. Trends Microbiol 22:113-119. 101. Johnston, J. W., D. E. Briles, L. E. Myers and S. K. Hollingshead. 2006. Mn2+-dependent regulation of multiple genes in Streptococcus pneumoniae through PsaR and the resultant impact on virulence. Infect Immun 74:1171-1180. 102. Jordan, H. T., C. L. Richards, Jr., D. C. Burton, M. C. Thigpen and C. A. Van Beneden. 2007. Group a streptococcal disease in long-term care facilities: descriptive epidemiology and potential control measures. Clin Infect Dis 45:742- 752. 103. Kadioglu, A., J. N. Weiser, J. C. Paton and P. W. Andrew. 2008. The role of Streptococcus pneumoniae virulence factors in host respiratory colonization and disease. Nat Rev Microbiol 6:288-301. 104. Kausmally, L., O. Johnsborg, M. Lunde, E. Knutsen and L. S. Havarstein. 2005. Choline-binding protein D (CbpD) in Streptococcus pneumoniae is essential for competence-induced cell lysis. J Bacteriol 187:4338-4345. 105. Kawai, F., M. Shoda, R. Harashima, Y. Sadaie, H. Hara and K. Matsumoto. 2004. Cardiolipin domains in Bacillus subtilis marburg membranes. J Bacteriol 186:1475-1483. 106. Keilin, D. 1959. The problem of anabiosis or latent life: history and current concept. Proc R Soc Lond B Biol Sci 150:149-191. 107. Kempf, B. and E. Bremer. 1998. Uptake and synthesis of compatible solutes as microbial stress responses to high-osmolality environments. Arch Microbiol 170:319-330. 169 108. Khroustalyova, G., J. Boudrant and A. Rapoport. 2009. Resistance of a recombinant Escherichia coli to dehydration. Cell Biol Int 33:1194-1195. 109. Kim, J. O. and J. N. Weiser. 1998. Association of intrastrain phase variation in quantity of capsular polysaccharide and teichoic acid with the virulence of Streptococcus pneumoniae. J Infect Dis 177:368-377. 110. Kim, J. S., H. S. Kim, J. Y. Park, H. S. Koo, C. S. Choi, W. Song, H. C. Cho and K. M. Lee. 2012. Contamination of X-ray cassettes with methicillin-resistant Staphylococcus aureus and methicillin-resistant Staphylococcus haemolyticus in a radiology department. Ann Lab Med 32:206-209. 111. Kim, P. E., D. M. Musher, W. P. Glezen, M. C. Rodriguez-Barradas, W. K. Nahm and C. E. Wright. 1996. Association of invasive pneumococcal disease with season, atmospheric conditions, air pollution, and the isolation of respiratory viruses. Clin Infect Dis 22:100-106. 112. Kramer, A. and O. Assadian. 2014. Survival of Microorganisms on Inanimate Surfaces. In G. Borkow (Ed.), Use of Biocidal Surfaces for Reduction of Healthcare Acquired Infections: 7-26. Springer, New York. 113. Kramer, A., I. Schwebke and G. Kampf. 2006. How long do nosocomial pathogens persist on inanimate surfaces? A systematic review. BMC Infect Dis 6:130. 114. Kultz, D. 2005. Molecular and evolutionary basis of the cellular stress response. Annu Rev Physiol 67:225-257. 115. Kung, C., B. Martinac and S. Sukharev. 2010. Mechanosensitive channels in microbes. Annu Rev Microbiol 64:313-329. 116. Kunkel, T. A. and D. A. Erie. 2005. DNA mismatch repair. Annu Rev Biochem 74:681-710. 117. Kusumaningrum, H. D., G. Riboldi, W. C. Hazeleger and R. R. Beumer. 2003. Survival of foodborne pathogens on stainless steel surfaces and cross- contamination to foods. Int J Food Microbiol 85:227-236. 118. Kwon, H. Y., S. W. Kim, M. H. Choi, A. D. Ogunniyi, J. C. Paton, S. H. Park, S. N. Pyo and D. K. Rhee. 2003. Effect of heat shock and mutations in ClpL and ClpP on virulence gene expression in Streptococcus pneumoniae. Infect Immun 71:3757-3765. 119. Lanie, J. A., W. L. Ng, K. M. Kazmierczak, T. M. Andrzejewski, T. M. Davidsen, K. J. Wayne, H. Tettelin, J. I. Glass and M. E. Winkler. 2007. Genome sequence of Avery's virulent serotype 2 strain D39 of Streptococcus pneumoniae and comparison with that of unencapsulated laboratory strain R6. J Bacteriol 189:38-51. 120. Laufer, A. S., J. P. Metlay, J. F. Gent, K. P. Fennie, Y. Kong and M. M. Pettigrew. 2011. Microbial communities of the upper respiratory tract and otitis media in children. MBio 2:e00245-00210. 121. Lazinski, D. W. and A. Camilli. 2013. Homopolymer tail-mediated ligation PCR: a streamlined and highly efficient method for DNA cloning and library construction. Biotechniques 54:25-34. 122. Lee, A. G. 2004. How lipids affect the activities of integral membrane proteins. Biochim Biophys Acta 1666:62-87. 170 123. Lee, L., S. Tin and S. T. Kelley. 2007. Culture-independent analysis of bacterial diversity in a child-care facility. BMC Microbiol 7:27. 124. Lemos, J. A., E. Tsakalidou and K. Papadimitriou. 2011. Stress Responses of Streptococci. In E. Tsakalidou and K. Papadimitriou (Eds.), Stress Responses of Lactic Acid Bacteria: 251-303. Springer, New York. 125. Levina, N., S. Totemeyer, N. R. Stokes, P. Louis, M. A. Jones and I. R. Booth. 1999. Protection of Escherichia coli cells against extreme turgor by activation of MscS and MscL mechanosensitive channels: identification of genes required for MscS activity. Embo J 18:1730-1737. 126. Li, H., A. Bhaskara, C. Megalis and M. L. Tortorello. 2012. Transcriptomic analysis of Salmonella desiccation resistance. Foodborne Pathog Dis 9:1143- 1151. 127. Lievense, L. C. and K. van 't Riet. 1994. Convective drying of bacteria. II. Factors influencing survival. Adv Biochem Eng Biotechnol 51:71-89. 128. Linke, K. and U. Jakob. 2003. Not every disulfide lasts forever: disulfide bond formation as a redox switch. Antioxid Redox Signal 5:425-434. 129. Liu, Y., R. Prasad, W. A. Beard, P. S. Kedar, E. W. Hou, D. D. Shock and S. H. Wilson. 2007. Coordination of steps in single-nucleotide base excision repair mediated by apurinic/apyrimidinic endonuclease 1 and DNA polymerase beta. J Biol Chem 282:13532-13541. 130. Lopez, C. S., A. F. Alice, H. Heras, E. A. Rivas and C. Sanchez-Rivas. 2006. Role of anionic phospholipids in the adaptation of Bacillus subtilis to high salinity. Microbiology 152:605-616. 131. Lorca, G. L., R. D. Barabote, V. Zlotopolski, C. Tran, B. Winnen, R. N. Hvorup, A. J. Stonestrom, E. Nguyen, L. W. Huang, D. S. Kim and M. H. Saier, Jr. 2007. Transport capabilities of eleven gram-positive bacteria: comparative genomic analyses. Biochim Biophys Acta 1768:1342-1366. 132. Lowen, A. C., S. Mubareka, J. Steel and P. Palese. 2007. Influenza virus transmission is dependent on relative humidity and temperature. PLoS Pathog 3:1470-1476. 133. Magee, A. D. and J. Yother. 2001. Requirement for capsule in colonization by Streptococcus pneumoniae. Infect Immun 69:3755-3761. 134. Marinella, M. A., C. Pierson and C. Chenoweth. 1997. The stethoscope. A potential source of nosocomial infection? Arch Intern Med 157:786-790. 135. Marks, L. R., B. A. Davidson, P. R. Knight and A. P. Hakansson. 2013. Interkingdom signaling induces Streptococcus pneumoniae biofilm dispersion and transition from asymptomatic colonization to disease. MBio 4: 136. Marks, L. R., G. I. Parameswaran and A. P. Hakansson. 2012. Pneumococcal interactions with epithelial cells are crucial for optimal biofilm formation and colonization in vitro and in vivo. Infect Immun 80:2744-2760. 137. Marks, L. R., R. M. Reddinger and A. P. Hakansson. 2012. High levels of genetic recombination during nasopharyngeal carriage and biofilm formation in Streptococcus pneumoniae. MBio 3: 138. Marks, L. R., R. M. Reddinger and A. P. Hakansson. 2014. Biofilm formation enhances fomite survival of Streptococcus pneumoniae and Streptococcus pyogenes. Infect Immun 82:1141-1146. 171 139. Marshall, B., R. A. Sen, P. R. Chadwick and M. G. Keaney. 1998. Environmental contamination of a new general surgical ward. J Hosp Infect 39:242-243. 140. Martinac, B., T. Nomura, G. Chi, E. Petrov, P. R. Rohde, A. R. Battle, A. Foo, M. Constantine, R. Rothnagel, S. Carne, E. Deplazes, B. Cornell, C. G. Cranfield, B. Hankamer and M. J. Landsberg. 2014. Bacterial mechanosensitive channels: models for studying mechanosensory transduction. Antioxid Redox Signal 20:952-969. 141. Mary, P. M., N.; Tailliez, R. 1993. Production and survival during storage of spray-dried Bradyrhizobium japonicum cell concentrates. J Appl Microbiol 74:340-344. 142. Mascher, T., M. Heintz, D. Zahner, M. Merai and R. Hakenbeck. 2006. The CiaRH system of Streptococcus pneumoniae prevents lysis during stress induced by treatment with cell wall inhibitors and by mutations in pbp2x involved in beta- lactam resistance. J Bacteriol 188:1959-1968. 143. Matsumoto, K., J. Kusaka, A. Nishibori and H. Hara. 2006. Lipid domains in bacterial membranes. Mol Microbiol 61:1110-1117. 144. McLaggan, D., J. Naprstek, E. T. Buurman and W. Epstein. 1994. Interdependence of K+ and glutamate accumulation during osmotic adaptation of Escherichia coli. J Biol Chem 269:1911-1917. 145. Michel, B. 2005. After 30 years of study, the bacterial SOS response still surprises us. PLoS Biol 3:e255. 146. Mileykovskaya, E. and W. Dowhan. 2009. Cardiolipin membrane domains in prokaryotes and eukaryotes. Biochim Biophys Acta 1788:2084-2091. 147. Mileykovskaya, E., Q. Sun, W. Margolin and W. Dowhan. 1998. Localization and function of early cell division proteins in filamentous Escherichia coli cells lacking phosphatidylethanolamine. J Bacteriol 180:4252-4257. 148. Miller, L. G. and B. A. Diep. 2008. Clinical practice: colonization, fomites, and virulence: rethinking the pathogenesis of community-associated methicillin- resistant Staphylococcus aureus infection. Clin Infect Dis 46:752-760. 149. Mitscherlich, E. and E. H. Marth. 1984. Microbial Survival in the Environment. Springer-Verlag, Berlin. 150. Miyamoto-Shinohara, Y., J. Sukenobe, T. Imaizumi and T. Nakahara. 2008. Survival of freeze-dried bacteria. J Gen Appl Microbiol 54:9-24. 151. Moore, G., I. S. Blair and D. A. McDowell. 2007. Recovery and transfer of Salmonella Typhimurium from four different domestic food contact surfaces. J Food Prot 70:2273-2280. 152. Moretro, T., E. Heir, K. R. Mo, O. Habimana, A. Abdelgani and S. Langsrud. 2010. Factors affecting survival of Shigatoxin-producing Escherichia coli on abiotic surfaces. Int J Food Microbiol 138:71-77. 153. Morimatsu, K. and S. C. Kowalczykowski. 2003. RecFOR proteins load RecA protein onto gapped DNA to accelerate DNA strand exchange: a universal step of recombinational repair. Mol Cell 11:1337-1347. 154. Munoz-Elias, E. J., J. Marcano and A. Camilli. 2008. Isolation of Streptococcus pneumoniae biofilm mutants and their characterization during nasopharyngeal colonization. Infect Immun 76:5049-5061. 172 155. Musher, D. M. 2003. How contagious are common respiratory tract infections? N Engl J Med 348:1256-1266. 156. Nakano, S., K. N. Erwin, M. Ralle and P. Zuber. 2005. Redox-sensitive transcriptional control by a thiol/disulphide switch in the global regulator, Spx. Mol Microbiol 55:498-510. 157. National Institute on Deafness and Other Communication Disorders and National Institutes of Health. 2010. NIDCD Fact Sheet: Ear Infections in Children. 1-6. 158. Neely, A. N. and M. P. Maley. 2000. Survival of enterococci and staphylococci on hospital fabrics and plastic. J Clin Microbiol 38:724-726. 159. Nelson, A. L., A. M. Roche, J. M. Gould, K. Chim, A. J. Ratner and J. N. Weiser. 2007. Capsule enhances pneumococcal colonization by limiting mucus- mediated clearance. Infect Immun 75:83-90. 160. Nguyen, D. M., L. Mascola and E. Brancoft. 2005. Recurring methicillin- resistant Staphylococcus aureus infections in a football team. Emerg Infect Dis 11:526-532. 161. Nishibori, A., J. Kusaka, H. Hara, M. Umeda and K. Matsumoto. 2005. Phosphatidylethanolamine domains and localization of phospholipid synthases in Bacillus subtilis membranes. J Bacteriol 187:2163-2174. 162. Novak, K. D., R. P. Kowalski, L. M. Karenchak and Y. J. Gordon. 1995. Chlamydia trachomatis can be transmitted by a nonporous plastic surface in vitro. Cornea 14:523-526. 163. Novakova, L., L. Saskova, P. Pallova, J. Janecek, J. Novotna, A. Ulrych, J. Echenique, M. C. Trombe and P. Branny. 2005. Characterization of a eukaryotic type serine/threonine protein kinase and protein phosphatase of Streptococcus pneumoniae and identification of kinase substrates. Febs J 272:1243-1254. 164. Nystrom, T. 2001. Not quite dead enough: on bacterial life, culturability, senescence, and death. Arch Microbiol 176:159-164. 165. Obaro, S. and R. Adegbola. 2002. The pneumococcus: carriage, disease and conjugate vaccines. J Med Microbiol 51:98-104. 166. Ogunniyi, A. D., L. K. Mahdi, M. P. Jennings, A. G. McEwan, C. A. McDevitt, M. B. Van der Hoek, C. J. Bagley, P. Hoffmann, K. A. Gould and J. C. Paton. 2010. Central role of manganese in regulation of stress responses, physiology, and metabolism in Streptococcus pneumoniae. J Bacteriol 192:4489- 4497. 167. Omelchenko, M. V., Y. I. Wolf, E. K. Gaidamakova, V. Y. Matrosova, A. Vasilenko, M. Zhai, M. J. Daly, E. V. Koonin and K. S. Makarova. 2005. Comparative genomics of Thermus thermophilus and Deinococcus radiodurans: divergent routes of adaptation to thermophily and radiation resistance. BMC Evol Biol 5:57. 168. Ophir, T. and D. L. Gutnick. 1994. A role for exopolysaccharides in the protection of microorganisms from desiccation. Appl Environ Microbiol 60:740- 745. 169. Otter, J. A., S. Yezil and G. L. French. 2014. The Role of Contaminated Surfaces in the Transmission of Nosocomial Pathogens. In G. Borkow (Ed.), Use 173 of Biocidal Surfaces for Reduction of Healthcare Acquired Infections: 27-58. Springer, New York. 170. Parer, S., A. Lotthe, P. Chardon, R. Poncet, H. Jean-Pierre and E. Jumas- Bilak. 2012. An outbreak of heterogeneous glycopeptide-intermediate Staphylococcus aureus related to a device source in an intensive care unit. Infect Control Hosp Epidemiol 33:167-174. 171. Pelton, S. I., H. Huot, J. A. Finkelstein, C. J. Bishop, K. K. Hsu, J. Kellenberg, S. S. Huang, R. Goldstein and W. P. Hanage. 2007. Emergence of 19A as virulent and multidrug resistant Pneumococcus in Massachusetts following universal immunization of infants with pneumococcal conjugate vaccine. Pediatr Infect Dis J 26:468-472. 172. Pembrey, R. S., K. C. Marshall and R. P. Schneider. 1999. Cell surface analysis techniques: What do cell preparation protocols do to cell surface properties? Appl Environ Microbiol 65:2877-2894. 173. Pericone, C. D., K. Overweg, P. W. Hermans and J. N. Weiser. 2000. Inhibitory and bactericidal effects of hydrogen peroxide production by Streptococcus pneumoniae on other inhabitants of the upper respiratory tract. Infect Immun 68:3990-3997. 174. Pericone, C. D., S. Park, J. A. Imlay and J. N. Weiser. 2003. Factors contributing to hydrogen peroxide resistance in Streptococcus pneumoniae include pyruvate oxidase (SpxB) and avoidance of the toxic effects of the fenton reaction. J Bacteriol 185:6815-6825. 175. Pesakhov, S., R. Benisty, N. Sikron, Z. Cohen, P. Gomelsky, I. Khozin- Goldberg, R. Dagan and N. Porat. 2007. Effect of hydrogen peroxide production and the Fenton reaction on membrane composition of Streptococcus pneumoniae. Biochim Biophys Acta 1768:590-597. 176. Peterson, S., R. T. Cline, H. Tettelin, V. Sharov and D. A. Morrison. 2000. Gene expression analysis of the Streptococcus pneumoniae competence regulons by use of DNA microarrays. J Bacteriol 182:6192-6202. 177. Pillai, D. R., D. Shahinas, A. Buzina, R. A. Pollock, R. Lau, K. Khairnar, A. Wong, D. J. Farrell, K. Green, A. McGeer and D. E. Low. 2009. Genome-wide dissection of globally emergent multi-drug resistant serotype 19A Streptococcus pneumoniae. BMC Genomics 10:642. 178. Pinto, A. N., T. Phan, G. Sala, E. Y. Cheong, S. Siarakas and T. Gottlieb. 2011. Reusable venesection tourniquets: a potential source of hospital transmission of multiresistant organisms. Med J Aust 195:276-279. 179. Potts, M. 1994. Desiccation tolerance of prokaryotes. Microbiol Rev 58:755-805. 180. Potts, M. 2001. Desiccation tolerance: a simple process? Trends Microbiol 9:553- 559. 181. Prudhomme, M., L. Attaiech, G. Sanchez, B. Martin and J. P. Claverys. 2006. Antibiotic stress induces genetic transformability in the human pathogen Streptococcus pneumoniae. Science 313:89-92. 182. Record, M. T., Jr., E. S. Courtenay, D. S. Cayley and H. J. Guttman. 1998. Responses of E. coli to osmotic stress: large changes in amounts of cytoplasmic solutes and water. Trends Biochem Sci 23:143-148. 174 183. Regev-Yochay, G., K. Trzcinski, C. M. Thompson, M. Lipsitch and R. Malley. 2007. SpxB is a suicide gene of Streptococcus pneumoniae and confers a selective advantage in an in vivo competitive colonization model. J Bacteriol 189:6532-6539. 184. Reitmeyer, J. C., A. Ewert, M. A. Crawford, G. R. Reitmeyer and L. Mock. 1993. Survival of group A streptococci in dried human blood. J Med Microbiol 38:61-63. 185. Reuter, M., N. J. Hayward, S. S. Black, S. Miller, D. T. Dryden and I. R. Booth. 2014. Mechanosensitive channels and bacterial cell wall integrity: does life end with a bang or a whimper? J R Soc Interface 11:20130850. 186. Reynolds, K. A. and I. L. Pepper. 2000. Microorganisms in the Environment. In R. M. Maier, I. L. Pepper and C. P. Gerba (Eds.), Environmental Microbiology: 7- 42. Academic Press, San Diego. 187. Rhee, S. G. 1999. Redox signaling: hydrogen peroxide as intracellular messenger. Exp Mol Med 31:53-59. 188. Rhee, S. G., T. S. Chang, Y. S. Bae, S. R. Lee and S. W. Kang. 2003. Cellular regulation by hydrogen peroxide. J Am Soc Nephrol 14:S211-215. 189. Roberson, E. B. and M. K. Firestone. 1992. Relationship between Desiccation and Exopolysaccharide Production in a Soil Pseudomonas sp. Appl Environ Microbiol 58:1284-1291. 190. Romantsov, T., Z. Guan and J. M. Wood. 2009. Cardiolipin and the osmotic stress responses of bacteria. Biochim Biophys Acta 1788:2092-2100. 191. Romantsov, T., S. Helbig, D. E. Culham, C. Gill, L. Stalker and J. M. Wood. 2007. Cardiolipin promotes polar localization of osmosensory transporter ProP in Escherichia coli. Mol Microbiol 64:1455-1465. 192. Rosch, J. W., G. Gao, G. Ridout, Y. D. Wang and E. I. Tuomanen. 2009. Role of the manganese efflux system mntE for signalling and pathogenesis in Streptococcus pneumoniae. Mol Microbiol 72:12-25. 193. Rose, L. J., R. Donlan, S. N. Banerjee and M. J. Arduino. 2003. Survival of Yersinia pestis on environmental surfaces. Appl Environ Microbiol 69:2166-2171. 194. Rosen, R. and E. Z. Ron. 2002. Proteome analysis in the study of the bacterial heat-shock response. Mass Spectrom Rev 21:244-265. 195. Ruotolo, B. T. and C. V. Robinson. 2006. Aspects of native proteins are retained in vacuum. Curr Opin Chem Biol 10:402-408. 196. Saleh, M., S. G. Bartual, M. R. Abdullah, I. Jensch, T. M. Asmat, L. Petruschka, T. Pribyl, M. Gellert, C. H. Lillig, H. Antelmann, J. A. Hermoso and S. Hammerschmidt. 2013. Molecular architecture of Streptococcus pneumoniae surface thioredoxin-fold lipoproteins crucial for extracellular oxidative stress resistance and maintenance of virulence. EMBO Mol Med 5:1852-1870. 197. Sarangi, J. and R. Rowsell. 1995. A nursing home outbreak of group A streptococcal infection: case control study of environmental contamination. J Hosp Infect 30:162-164. 198. Saskova, L., L. Novakova, M. Basler and P. Branny. 2007. Eukaryotic-type serine/threonine protein kinase StkP is a global regulator of gene expression in Streptococcus pneumoniae. J Bacteriol 189:4168-4179. 175 199. Schlame, M. 2008. Cardiolipin synthesis for the assembly of bacterial and mitochondrial membranes. J Lipid Res 49:1607-1620. 200. Schmid-Hempel, P. and S. A. Frank. 2007. Pathogenesis, virulence, and infective dose. PLoS Pathog 3:1372-1373. 201. Schweder, T. and M. Hecker. 2004. Monitoring of Stress Responses. In S. O. Enfors (Ed.), Physiological Stress Responses in Bioprocesses: 47-71. Springer- Verlag, Berlin. 202. Scott, E. and S. F. Bloomfield. 1990. The survival and transfer of microbial contamination via cloths, hands and utensils. J Appl Bacteriol 68:271-278. 203. Setlow, P. 1995. Mechanisms for the prevention of damage to DNA in spores of Bacillus species. Annu Rev Microbiol 49:29-54. 204. Sghaier, H., I. Narumi, K. Satoh, H. Ohba and H. Mitomo. 2007. Problems with the current deinococcal hypothesis: an alternative theory. Theory Biosci 126:43-45. 205. Shainheit, M. G., M. Mule and A. Camilli. 2014. The core promoter of the capsule operon of Streptococcus pneumoniae is necessary for colonization and invasive disease. Infect Immun 82:694-705. 206. Sherertz, R. J. and M. L. Sullivan. 1985. An outbreak of infections with Acinetobacter calcoaceticus in burn patients: contamination of patients' mattresses. J Infect Dis 151:252-258. 207. Shoemaker, N. B. and W. R. Guild. 1974. Destruction of low efficiency markers is a slow process occurring at a heteroduplex stage of transformation. Mol Gen Genet 128:283-290. 208. Singleton, M. R., M. S. Dillingham, M. Gaudier, S. C. Kowalczykowski and D. B. Wigley. 2004. Crystal structure of RecBCD enzyme reveals a machine for processing DNA breaks. Nature 432:187-193. 209. Slade, D. and M. Radman. 2011. Oxidative Stress Resistance in Deinococcus radiodurans. Microbiol Mol Biol Rev 75:133-191. 210. Spellerberg, B., D. R. Cundell, J. Sandros, B. J. Pearce, I. Idanpaan- Heikkila, C. Rosenow and H. R. Masure. 1996. Pyruvate oxidase, as a determinant of virulence in Streptococcus pneumoniae. Mol Microbiol 19:803- 813. 211. Stacey, A., P. Burden, C. Croton and E. Jones. 1998. Contamination of television sets by methicillin-resistant Staphylococcus aureus (MRSA). J Hosp Infect 39:243-244. 212. Stasic, A. J., A. C. Lee Wong and C. W. Kaspar. 2012. Osmotic and desiccation tolerance in Escherichia coli O157:H7 requires rpoS (sigma(38)). Curr Microbiol 65:660-665. 213. Steil, L., T. Hoffmann, I. Budde, U. Volker and E. Bremer. 2003. Genome- wide transcriptional profiling analysis of adaptation of Bacillus subtilis to high salinity. J Bacteriol 185:6358-6370. 214. Steinmoen, H., A. Teigen and L. S. Havarstein. 2003. Competence-induced cells of Streptococcus pneumoniae lyse competence-deficient cells of the same strain during cocultivation. J Bacteriol 185:7176-7183. 215. Stoodley, P., K. Sauer, D. G. Davies and J. W. Costerton. 2002. Biofilms as complex differentiated communities. Annu Rev Microbiol 56:187-209. 176 216. Sukharev, S. 2002. Purification of the small mechanosensitive channel of Escherichia coli (MscS): the subunit structure, conduction, and gating characteristics in liposomes. Biophys J 83:290-298. 217. Tacconelli, E. 2011. When did the doctors become fomites? Clin Microbiol Infect 17:794-796. 218. Tang, J. W. 2009. The effect of environmental parameters on the survival of airborne infectious agents. J R Soc Interface 6 Suppl 6:S737-746. 219. Teng, S. O., W. S. Lee, T. Y. Ou, Y. C. Hsieh, W. C. Lee and Y. C. Lin. 2009. Bacterial contamination of patients' medical charts in a surgical ward and the intensive care unit: impact on nosocomial infections. J Microbiol Immunol Infect 42:86-91. 220. Tettelin, H., K. E. Nelson, I. T. Paulsen, J. A. Eisen, T. D. Read, S. Peterson, J. Heidelberg, R. T. DeBoy, D. H. Haft, R. J. Dodson, A. S. Durkin, M. Gwinn, J. F. Kolonay, W. C. Nelson, J. D. Peterson, L. A. Umayam, O. White, S. L. Salzberg, M. R. Lewis, D. Radune, E. Holtzapple, H. Khouri, A. M. Wolf, T. R. Utterback, C. L. Hansen, L. A. McDonald, T. V. Feldblyum, S. Angiuoli, T. Dickinson, E. K. Hickey, I. E. Holt, B. J. Loftus, F. Yang, H. O. Smith, J. C. Venter, B. A. Dougherty, D. A. Morrison, S. K. Hollingshead and C. M. Fraser. 2001. Complete genome sequence of a virulent isolate of Streptococcus pneumoniae. Science 293:498-506. 221. Tolba, O., A. Loughrey, C. E. Goldsmith, B. C. Millar, P. J. Rooney and J. E. Moore. 2007. Survival of epidemic strains of nosocomial- and community- acquired methicillin-resistant Staphylococcus aureus on coins. Am J Infect Control 35:342-346. 222. Tran, T. D., H. Y. Kwon, E. H. Kim, K. W. Kim, D. E. Briles, S. Pyo and D. K. Rhee. 2011. Decrease in penicillin susceptibility due to heat shock protein ClpL in Streptococcus pneumoniae. Antimicrob Agents Chemother 55:2714- 2728. 223. Tran, T. D., H. Y. Kwon, E. H. Kim, K. W. Kim, D. E. Briles, S. Pyo and D. K. Rhee. 2011. Heat-shock protein ClpL/HSP100 increases penicillin tolerance in Streptococcus pneumoniae. Adv Otorhinolaryngol 72:126-128. 224. Truglio, J. J., D. L. Croteau, B. Van Houten and C. Kisker. 2006. Prokaryotic nucleotide excision repair: the UvrABC system. Chem Rev 106:233-252. 225. Tseng, H. J., A. G. McEwan, J. C. Paton and M. P. Jennings. 2002. Virulence of Streptococcus pneumoniae: PsaA mutants are hypersensitive to oxidative stress. Infect Immun 70:1635-1639. 226. Turlan, C., M. Prudhomme, G. Fichant, B. Martin and C. Gutierrez. 2009. SpxA1, a novel transcriptional regulator involved in X-state (competence) development in Streptococcus pneumoniae. Mol Microbiol 73:492-506. 227. Tzeng, Y. L., L. E. Martin and D. S. Stephens. 2014. Environmental survival of Neisseria meningitidis. Epidemiol Infect 142:187-190. 228. Ulger, F., S. Esen, A. Dilek, K. Yanik, M. Gunaydin and H. Leblebicioglu. 2009. Are we aware how contaminated our mobile phones with nosocomial pathogens? Ann Clin Microbiol Antimicrob 8:7. 229. van der Poll, T. and S. M. Opal. 2009. Pathogenesis, treatment, and prevention of pneumococcal pneumonia. Lancet 374:1543-1556. 177 230. van Opijnen, T., K. L. Bodi and A. Camilli. 2009. Tn-seq: high-throughput parallel sequencing for fitness and genetic interaction studies in microorganisms. Nat Methods 6:767-772. 231. van Opijnen, T. and A. Camilli. 2010. Genome-wide fitness and genetic interactions determined by Tn-seq, a high-throughput massively parallel sequencing method for microorganisms. Curr Protoc Microbiol Chapter 1:Unit1E 3. 232. van Opijnen, T. and A. Camilli. 2012. A fine scale phenotype-genotype virulence map of a bacterial pathogen. Genome Res 22:2541-2551. 233. Verhagen, L. M., M. I. de Jonge, P. Burghout, K. Schraa, L. Spagnuolo, S. Mennens, M. J. Eleveld, C. E. van der Gaast-de Jongh, A. Zomer, P. W. Hermans and H. J. Bootsma. 2014. Genome-wide identification of genes essential for the survival of Streptococcus pneumoniae in human saliva. PLoS One 9:e89541. 234. Walker, J. A., R. L. Allen, P. Falmagne, M. K. Johnson and G. J. Boulnois. 1987. Molecular cloning, characterization, and complete nucleotide sequence of the gene for pneumolysin, the sulfhydryl-activated toxin of Streptococcus pneumoniae. Infect Immun 55:1184-1189. 235. Walsh, R. L. and A. Camilli. 2011. Streptococcus pneumoniae is desiccation tolerant and infectious upon rehydration. MBio 2:e00092-00011. 236. Walther, B. A. and P. W. Ewald. 2004. Pathogen survival in the external environment and the evolution of virulence. Biol Rev Camb Philos Soc 79:849- 869. 237. Watson, S. P., M. O. Clements and S. J. Foster. 1998. Characterization of the starvation-survival response of Staphylococcus aureus. J Bacteriol 180:1750- 1758. 238. Weber, D. J. and W. A. Rutala. 1997. Role of environmental contamination in the transmission of vancomycin-resistant enterococci. Infect Control Hosp Epidemiol 18:306-309. 239. Weiser, J. N. 2010. The pneumococcus: why a commensal misbehaves. J Mol Med 88:97-102. 240. Weiser, J. N., R. Austrian, P. K. Sreenivasan and H. R. Masure. 1994. Phase variation in pneumococcal opacity: relationship between colonial morphology and nasopharyngeal colonization. Infect Immun 62:2582-2589. 241. Walther, B. A. and P. W. Ewald. 2004. Pathogen survival in the external environment and the evolution of virulence. Biol Rev Camb Philos Soc 79:849- 869. 242. Whatmore, A. M. and R. H. Reed. 1990. Determination of turgor pressure in Bacillus subtilis: a possible role for K+ in turgor regulation. J Gen Microbiol 136:2521-2526. 243. White, A. N., V. Ng, C. V. Spain, C. C. Johnson, L. M. Kinlin and D. N. Fisman. 2009. Let the sun shine in: effects of ultraviolet radiation on invasive pneumococcal disease risk in Philadelphia, Pennsylvania. BMC Infect Dis 9:196. 244. Whittington, A. M., G. Whitlow, D. Hewson, C. Thomas and S. J. Brett. 2009. Bacterial contamination of stethoscopes on the intensive care unit. Anaesthesia 64:620-624. 178 245. Wood, J. M. 2007. Bacterial osmosensing transporters. Methods Enzymol 428:77-107. 246. World Health Organization. 1999. Pneumococcal Vaccines. WHO Position Paper. Weekly Epidemiological Record 74:177-184. 247. World Health Organization. 2003. Pneumococcal Vaccines. WHO Position Paper. Weekly Epidemiological Record 78:97-120. 248. World Health Organization. 2012. Pneumococcal Vaccines. WHO Position Paper. Weekly Epidemiological Record 87:129-144. 249. Wu, H. Y., A. Virolainen, B. Mathews, J. King, M. W. Russell and D. E. Briles. 1997. Establishment of a Streptococcus pneumoniae nasopharyngeal colonization model in adult mice. Microb Pathog 23:127-137. 250. Xie, X., Y. Li, T. Zhang and H. H. Fang. 2006. Bacterial survival in evaporating deposited droplets on a teflon-coated surface. Appl Microbiol Biotechnol 73:703- 712. 251. Yan, D., T. P. Ikeda, A. E. Shauger and S. Kustu. 1996. Glutamate is required to maintain the steady-state potassium pool in Salmonella Typhimurium. Proc Natl Acad Sci U S A 93:6527-6531. 252. Yershov, A. L., B. S. Jordan, C. H. Guymon and M. A. Dubick. 2005. Relationship between the inoculum dose of Streptococcus pneumoniae and pneumonia onset in a rabbit model. Eur Respir J 25:693-700. 253. Yesilkaya, H., A. Kadioglu, N. Gingles, J. E. Alexander, T. J. Mitchell and P. W. Andrew. 2000. Role of manganese-containing superoxide dismutase in oxidative stress and virulence of Streptococcus pneumoniae. Infect Immun 68:2819-2826. 254. Yother, J. 2011. Capsules of Streptococcus pneumoniae and other bacteria: paradigms for polysaccharide biosynthesis and regulation. Annu Rev Microbiol 65:563-581. 255. Youngster, I., M. Berkovitch, E. Heyman, Z. Lazarovitch and M. Goldman. 2008. The stethoscope as a vector of infectious diseases in the paediatric division. Acta Paediatr 97:1253-1255. 256. Zimbudzi, E., R. L. Stuart, T. M. Korman and D. Kotsanas. 2011. Contamination of renal patients' hospital chart covers with vancomycin-resistant enterococci: Handle with care. Australas Med J 4:538-541. 179