Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Methicillin-resistant Staphylococcus aureus: molecular epidemiology, virulence in disease, and antibiotic modulation of virulence
(USC Thesis Other)
Methicillin-resistant Staphylococcus aureus: molecular epidemiology, virulence in disease, and antibiotic modulation of virulence
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Methicillin-Resistant Staphylococcus aureus:
Molecular Epidemiology, Virulence in Disease,
and Antibiotic Modulation of Virulence
by
Jason Yamaki
A Dissertation
Presented to the Faculty of The University of Southern California Graduate School
In Partial Fulfillment of Requirements
For the Degree
Doctor of Philosophy
PI and Advisor: Annie Wong-Beringer, Pharm.D.
Co-Advisor: Kathleen Rodgers, Ph.D.
August 2013
ii
Acknowledgements
I owe my deepest gratitude to my advisor Dr. Annie Wong-Beringer for her clinical insight,
encouragement, support, and the opportunity to begin doing research as a Pharm.D. student,
without this opportunity I would not have found an interest in research.
I sincerely thank my co-advisor Dr. Kathleen Rodgers who has supported me in my basic science
research and animal model of infection. I also wish to thank my dissertation committee members
Dr. Ronald Alkana and Dr. Andre Ouellette. And Drs. Paul Beringer and Roger Duncan who
have provided assistance throughout both my Ph.D. and Pharm.D. studies, and participated in
extracurricular tennis activities as stress relief.
I also thank all of my laboratory lab-mates, Tim and Joyce Bensman, Melissa Coyle, Henry Ho,
Albert Nguyen, and Ronald Sim for not only assisting me in experiments and teaching me new
techniques, but also for being such great friends.
I also wish to thank my family, my uncle Alex, and especially my wife Catherine Jabagat-
Yamaki, for giving me constant support throughout all of my studies.
Finally, I dedicate this dissertation to my grandparents Candelaria and Leo Ortega, whose roles
in my life was, and remains immeasurable.
iii
Abstract
Methicillin-resistant Staphylococcus aureus (MRSA) has become a major public health
problem due to the high prevalence of infections caused by these strains in the U.S. and
worldwide. MRSA is now the leading cause of skin soft tissue infections presented to emergency
departments (Moran et al., 2006) and a cause of 94,000 estimated cases of invasive infections per
year in the U.S. (Klevens et al., 2007). Adding to this challenge is the emergence of bacterial
strains resistant to standard treatment and unique virulent clones that affect otherwise healthy
individuals in the community. Molecular epidemiologic studies have described increasing
involvement of MRSA strains originated from the community (CA) as the cause of healthcare-
associated (HA) infections in other parts of the country.
The main focus of the studies described in this thesis was to explore the epidemiology of
MRSA strains within our institution and identify bacterial or patient markers that can be used to
guide treatment and improve outcomes. Furthermore, we investigated whether observations
described in other bacterial toxin-mediated disease is applicable to MRSA infection, including
correlation of in-vitro exotoxin production to disease severity and treatment approaches to inhibit
toxin production.
We found significant changes in the molecular epidemiology of MRSA within our
institution. Based on this change we offer possible markers that can be used to guide empiric
vancomycin treatment. We also present a timeframe by which patient response should be
evaluated and the possibility of switching from vancomycin to an alternative agent. Furthermore,
in our studies we found the in-vitro toxin production and decreased vancomycin susceptibility
did not have major impacts on disease severity and poor outcomes. Finally, we found that under
sub-inhibitory concentrations of protein synthesis inhibiting antibiotics PSM peptides can be
induced contrary to previous studies examining other exotoxins. This data challenges the
assumption that protein synthesis inhibiting antibiotics used in the treatment of MRSA infections
will only add additional benefits with no risks.
iv
This thesis is based on the following manuscripts
1. Yamaki J, Lee M, Shriner K, Wong-Beringer A. Can clinical and molecular
epidemiologic parameters guide empiric treatment with vancomycin for methicillin-
resistant Staphylococcus aureus infections? Diagn Microbiol Infect Dis 2011;70:124-130
2. Yamaki J, Synold T, Wong-Beringer A. Anti-virulence potential of TR-700 and
clindamycin on clinical isolates of Staphylococcus aureus producing phenol-soluble
modulins. Antimicrob Agents Chemother. 2011 Sep;55(9):4432-5
3. Yamaki J, Synold T, Wong-Beringer A. Virulence factor inhibition by linezolid,
tigecycline, and clindamycin against isolates causing invasive infections. Antimicrob
Agents Chemother 2013 (Conditionally accepted, April 2013)
4. Yamaki J, Minejima E, Nieberg P, Wong-Beringer A. Molecular epidemiology and in-
vitro production of protein A and alpha-hemolysin amongst MRSA strains causing
community-onset pneumonia. (In preparation)
5. Joo J, Yamaki J, Lou M, Hshieh S, Chu T, Shriner K, Wong-Beringer A. Early response
assessment to guide therapy for methicillin-resistant Staphylococcus aureus bacteremia.
Clin Therapeutics 2013 (Conditionally Accepted, May 2013)
v
Table of Contents
PAGE
ACKNOWLEDGEMENTS...……………………………………………………………………….….......ii
ABSTRACT…………………………………………………………………………………….…….…...iii
PREFACE……………………………………………………………………………………….……….....v
LIST OF ABBREVIATIONS……………………………………………………………………………...vi
LIST OF MULTIMEDIA OBJECTS ……………………………………………………………….….viii
CHAPTERS
CHAPTER 1 – Introduction and significance………………………………………………………….......1
CHAPTER 2 – Clinical and molecular epidemiology of MRSA infections and associated
outcomes……….………………………………………….………………………………24
CHAPTER 3 – Contribution of resistance and exoprotein production by MRSA strains to severity of
infection and outcomes…………………………………………………………………....49
CHAPTER 4 – Effects of protein synthesis inhibitors on exotoxin modulation in MRSA, in-vitro and in a
cellular model of infection …………………………………………………………..........68
CHAPTER 5 – Summary and future directions………...…………………………………………….…...96
REFERENCES………………………………………………………………………………….………..102
vi
Preface
The aim of this dissertation is to highlight the importance of CA-MRSA as a pathogen in skin
soft tissue infections and in hospitalized patients with invasive disease, identify bacterial or
patient markers that can be used to guide treatment, and the effects of antibiotics at sub-
inhibitory concentrations on MRSA exotoxin production. Based on patient clinical data,
molecular methods of genotyping, and assays quantifying toxin production we found CA-MRSA
strains are responsible for a significant proportion of hospitalizations due to Staphylococcus
aureus in geographic region of Los Angeles County and that the general assumption that protein
synthesis inhibiting antibiotics will inhibit exotoxin production at sub-inhibitory concentrations
cannot be universally applied.
vii
Abbreviations Used
Abbreviations
agr Accessory gene regulator
AMP Antimicrobial peptide
ANOVA Analysis of Variance
APACHE Acute Physiology and Chronic Health Evaluation
BAL Bronchoaveolar lavage
CA Community-associated
CAP Community-acquired pneumonia
CDC Centers for Disease Control
CFU Colony forming unit
CL Clindamycin
CLSI Clinical laboratory standards institute
COPD Chronic obstructive pulmonary disease
cSSSI Complicated skin skin structure infection
dNTP Deoxyribonucleotide
h Hour
HA Healthcare-associated
HCAP Health-care associated pneumonia
Hla α-hemolysin
HRP Horse radish peroxidase
hVISA Heteroresistant vancomycin intermediate Staphylococcus aureus
IBW Ideal body weight
ICU Intensive care unit
IL-17 Interleukin – 17
IL-1β Interleukin – 1 beta
IL-6 Interleukin – 6
IL-8 Interleukin – 8
IFN-γ Interferon gamma
IQR Interquartile range
LAC Los Angeles County
LC Liquid chromatography
LDH Lactic acid dehydrogenase
LOS Length of stay
LZ Linezolid
MIC Minimum inhibitory concentration
MLST Multi locus sequence typing
mRNA Messenger ribonucleic acid
MRSA Methicillin-resistant Staphylococcus aureus
MS Mass spectometry
MSSA Methicillin susceptible Staphylococcus aureus
NaCl Sodium chloride
NARSA Network on Antimicrobial Resistance in Staphylococcus aureus
NOD2 Nucleotide-binding oligomerization domain 2
OD Optical density
viii
Abbreviations Continued
Abbreviations
PBP Penicillin binding protein
PBS Phosphate buffer solution
PCR Polymerase chain reaction
PFGE Pulse field gel electrophoresis
PMN Polymorphonuclear leukocyte (Neutrophil)
pna Pneumonia
PRRs Pathogen Recognition Receptors
PSI Pneumonia severity index
PSM Phenol-soluble modulins
PVL Panton-Valentine leukocidin
RBC Red blood cell
RNA III Ribonucleic acid III
rRNA Ribosomal ribonucleic acid
RT-PCR Real time polymerase chain reaction
SaeRS Staphylococcal accessory element
SarA Staphylococcus accessory regulator A
SCCmec Staphylococcus Cassette Chromosome mec
SD Standard deviation
Spa Protein A
SSRs Short sequence repeats
SSTI Skin soft tissue infection
TNF Tumor necrosis factor
TNFR1 Tumor necrosis factor receptor
tRNA Transfer ribonucleic acid
TSB Tryptic soy broth
TYG Tigecycline
VAN Vancomycin
VAP Ventilator-associated pneumonia
ix
List of Multimedia Objects
Chapter 1:
Diagram 1. Diagram 1. Regulation of the agr system………………………………………….23
Chapter 2:
Table 1. PCR primers for SCCmec typing and detection of PVL……………………………....29
Table 2. Demographics, characteristics, and spectrum of disease of patients infected with
MRSA strains from the 2005 – 2007 cohort …………………………………………………....32
Table 3. Patient characteristics and outcomes in the bacteremia cohort…………………….….36
Table 4. Demographics, patient characteristics, and disease spectrum of strains with low or
high vancomycin MIC..................................................................................................................42
Figure 1. Distribution of PVL (+) MRSA strains among patient meeting clinical criteria for CA
or HA-MRSA as defined by CDC ...............................................................................................33
Figure 2. Distribution of source of MRSA bacteremia ...............................................................35
Figure 3. Distribution of Vancomycin MIC by SCCmec Type among strain isolated from
bacteremia patients........................................................................................................................43
Chapter 3:
Table 1. Disease severity and outcomes based on Hla and Spa production…………………….59
Table 2. Comparative susceptibility of MRSA strains to vancomycin, daptomycin, linezolid, and
tigecycline ....................................................................................................................................63
Figure 1. PSMα peptide production among Staphylococcus aureus clinical isolates causing
cSSSI................................................................ ............................................................................55
x
Figure 2. Variation in PSMα peptide production by PVL status and invasiveness…………….56
Figure 3. In-vitro α-hemolysin and protein A production based on PVL status………………..58
Figure 4. Differential cytotoxicity of clinical MRSA strains against A549
and human PMNs..........................................................................................................................60
Chapter 4:
Diagram 1. Improved lux reporter for use in Staphylococcus aureus.............................................71
Table 1. Primers used for RT-PCR and cloning...............................................................................74
Table 2. Baseline production of PSMα peptides by representative isolates selected for testing with
antibiotics..........................................................................................................................................79
Table 3. Toxin-specific responses to tigecycline exposure at ⅛ MIC in
representative strains.........................................................................................................................82
Figure 1. Growth of control strain LAC under antibiotic sub-inhibitory concentrations of
antibiotics..........................................................................................................................................76
Figure 2. Effects of TR-700 on PSMα
1-4
production.......................................................................77
Figure 3. Overall effects of sub-inhibitory concentrations of antibiotics on
PSMα
1-4
production................................................................ ..........................................................78
Figure 4. Strain-specific effects observed with tigecycline at ⅛ MIC on PSMα
1
production among
11 of clinical and two control isolates tested…………………………………................................80
Figure 5. Strain-specific effects observed with tigecycline at ⅛ MIC on PSMα
1
production among
11 of clinical and two control isolates tested……………………………………………………... 81
Figure 6. Effects of linezolid, clindamycin, and tigecycline on Hla production ……...………….84
Figure 7. Effects of linezolid, clindamycin, and tigecycline on Spa production………………….84
Figure 8. Effects of linezolid and tigecycline on exotoxin mRNA levels ………………….…….86
xi
Figure 9. psmα promoter activity as measure by luminescence …………………………………..87
Figure 10. Changes in global regulators of exotoxin production under antibiotic pressure…..….88
Figure 11. Cytotoxicity against A549 cells under sub-MIC tigecycline ……………….…...….…90
Figure 12. Cytotoxicity against human PMNs under sub-MIC tigecycline ………………………90
1
Chapter 1.
Background and Significance
S. aureus is a medically important bacteria that is readily adaptable to various environments and
conditions including antibiotic pressure, produces an arsenal of virulence factors it is not only a
major cause of hospital infections but now affects individuals in the community settings as well.
Methicillin-resistant S. aureus (MRSA) has now become a major public health concern causing a
wide spectrum of infections across healthcare and community settings, from localized skin
infections to life-threatening pneumonia (Klevens et al., 2007; Styers et al., 2006). It is the
leading cause of skin and skin structure infections (SSSIs) and a leading pathogen in invasive
infections including ventilator associated pneumonia (Tacconelli, 2009). In the U.S., deaths from
invasive MRSA infections exceeds those from AIDS annually (Styers et al., 2006).
Methicillin Resistance.
During the late 1950s, methicillin was introduced to combat penicillinase-producing S. aureus
strains. However, shortly after its introduction MRSA strains emerged and began infecting
patients within the hospital settings. Methicillin resistance is conferred by the mecA gene, which
encodes for the alternative penicillin-binding protein (PBP) known as PBP2a that exhibits a
1000-fold decrease in affinity for most beta-lactam antibiotics (Fuda et al., 2004; Pinho et al.,
2001). The Staphylococcal Cassette Chromosome (SCC) is a mobile genetic element that carries
the mecA gene and is always found adjacent to the oriC of the Staphylococcal chromosome
(Hiramatsu et al., 2004). The SCCmec element carries specific sets of recombinases that are
required for excision and integration of the element, and can carry additional antibiotic resistance
2
genes. Currently there are seven known SCCmec types, with the most prevalent types being I-V
in the U.S. (Ito et al., 2009).
The first SCCmec type is thought to be SCCmec I, which can be found in the laboratory
control strain COL, and was originally isolated in a hospitalized patient when MRSA first
appeared in the 1960s (Malachowa et al., 2012). SCCmec I - III are commonly found in HA-
MRSA strains and are larger compared to other types allowing these elements to carry additional
antibiotic resistance genes. SCCmec IV and V are commonly found in CA-MRSA strains and
infrequently carry additional resistance genes, leading to the commonly observed CA-MRSA
phenotype of being susceptible to non-beta-lactam antibiotics. Currently, SCCmec typing is a
simple and rapid method for use in epidemiology studies of CA and HA-MRSA.
Epidemiology and Strain Characterization.
As MRSA is a versatile organism, the epidemiology of infections has reportedly been changing
in other areas of the U.S. and other countries (Gonzalez et al., 2006; Maree et al., 2007).
Historically, MRSA infections were primarily restricted to the healthcare setting where patients
were most likely to encounter the organism. However, within the last two decades the sources
and places of acquisition began changing as MRSA strains began causing infections in the
community setting in persons who were otherwise healthy with no recent healthcare exposure
(Gillet et al., 2002). This shift caused the CDC to develop clinical criteria to determine whether
the infection was nosocomial in origin or community, leading to the classification of two clinical
infection types: healthcare-associated (HA) and community-associated (CA). Strains are
designated by CDC clinical definitions to be HA if the strain is isolated from the patient after
48h of admission, the patient has had healthcare exposure within the past year, or has indwelling
3
medical devices. The strain is considered CA if it is isolated within 48h of admission and the
patient has no prior medical exposure within the past year.
Upon further investigation, a particular set of MRSA clones was found to be the primary
cause of the new manifestation of CA infections. These clones primarily belong to the USA300
ST8 and USA400 ST1 clonal complex, although other clonal complexes exist within CA-MRSA
strains. These clones were primarily responsible for causing skin soft tissue infections (SSTIs).
However, severe necrotizing pneumonia by CA-MRSA strain MW2 (USA400) was reported to
be the cause of death in three children within the community setting in the late 1990s, with
additional necrotizing pneumonia cases being reported by strains of the same background. Since
the emergence of CA-MRSA strains, USA300 has now become the dominant CA strain in the
U.S. (Johnson et al., 2007).
Based on molecular characteristics, the two MRSA groups are known to harbor different
pathogenicity islands, prophages, antibiotic resistance genes, and polymorphisms in the
accessory gene regulator (agr), one of the global regulators of virulence. Nearly all (95%) of
CA-MRSA strains possess a prophage that encodes for the Panton Valentine leukocidin (PVL),
are either SCCmec type IV or V, and usually belong to agr groups I or III (Diep and Otto, 2008;
Tenover et al., 2006). On the other hand, HA-MRSA strains are typically SCCmec type I-III, do
not carry the PVL prophage, and usually are agr group II. However, both CA- and HA-MRSA
strains produce a common set of proteins associated with virulence that are encoded within the
core genome of nearly all S. aureus strains. Specifically, phenol soluble modulins (PSMs),
protein A (Spa), and α-hemolysin (Hla) are produced by nearly all S. aureus strains regardless of
methicillin resistance phenotype or background lineage (Li et al., 2009; Queck et al., 2008). It
has been suggested that one reason for the observed increase in virulence among CA-MRSA
4
strains compared to HA-MRSA strains is the increased production of Hla and PSMs encoded by
genes in the core genome and to a lesser extent mobile genetic elements encoding for the Panton
Valentine leukocidin (PVL) (Li et al., 2009).
As CA-MRSA infections have been known to cause severe necrotizing pneumonia and
necrotizing fasciitis (Gillet et al., 2002), it has been suggested that these CA-MRSA strains may
be more virulent than their HA counterparts (Mongkolrattanothai et al., 2003). Furthermore, a
few investigators have observed that CA-MRSA strains have increasingly become the cause of
HA infections, though the change in epidemiology may be geographically specific (Khatib et al.,
2009; Seybold et al., 2008; Zervos et al., 2008). In the face of this possible changing
epidemiology, an initial and necessary step to designing rational therapeutics to combat this
pathogen is to gain a better understanding of the clinical and molecular epidemiology of MRSA
strains in disease (Lina et al., 1998).
Contribution of CA-MRSA in hospitalized patients.
While CA-MRSA strains cause skin and soft tissue infections with or without abscess most
commonly, a large population-based study conducted by the CDC reported that up to 29% of
invasive MRSA infections were caused by the predominant USA300 clone of community origin.
Others have reported that nearly half of MRSA infections among hospitalized patients were CA
(Chua et al., 2008; Davis et al., 2006; Klevens et al., 2007). Two invasive MRSA infections of
major concern are bacteremia and pneumonia (Klevens et al., 2007). Bacterial persistence and
treatment failure in patients with MRSA bacteremia is increasingly reported and is believed to be
associated with vancomycin tolerance among infecting strains (Lodise et al., 2008; Sakoulas et
al., 2004; Soriano et al., 2008). Similarly, in patients with HA pneumonia (HAP), ventilator
5
associated pneumonia (VAP), and healthcare associated pneumonia (HCAP), higher mortality
rates and poor outcomes have been associated with MRSA strains with high vancomycin
minimum inhibitory concentrations (MIC) (Haque et al., 2010). The aforementioned studies
describe the association of outcomes based on vancomycin treatment. However, the contribution
of CA-MRSA in causing these infections is not well known as only a few studies have described
their association, with relatively low numbers of CA-MRSA strains found. Thus, the incidence
and outcomes related to these strains in bacteremia and pneumonia is relatively unknown.
In 2002 Gillet et al reported 16 cases of CA pneumonia caused by CA-MRSA strains
among French institutions from 1986-1998 (Gillet et al., 2002). Their goal was to compare the
outcomes associated with these strains that were defined as CA-MRSA because of the presence
of PVL and SCCmec IV, to those patients that were infected with PVL (-) strains. The
investigators found a highly significant (p=0.005) association with PVL (+) strains and risk for
mortality compared to patients with PVL (-) strains. Studies involving larger sample size here in
the U.S. did not report similar findings. Fowler et. al. collected 287 isolates of S. aureus in
patients with HAP; of those, only 8% were PVL (+) (Sharma-Kuinkel et al., 2012). No
association with poor outcomes and in-vitro production of PVL, Hla, or presence of other
virulence factors was found (Sharma-Kuinkel et al., 2012). Peyrani et al conducted a similar
study in HAP patients but limited it to MRSA isolates specifically. This study collected 109
isolates and 27% were PVL (+), however they too found no association in disease severity and
outcome (Peyrani et al., 2011).
While the above studies have found no association with virulence characteristics of the
causative strains (e.g. PVL (+), Hla production) it is important to note that the number of CA-
MRSA strains within each study were relatively small, and that these studies focused solely on
6
healthcare associated pneumonias. Seeing that the prevalence of CA-MRSA infections continues
to rise (Hadler et al., 2012), the contribution CA-MRSA strains and in-vitro production of key
virulence factors in community-associated and community-onset pneumonia will need to be
addressed.
Virulence Determinants in MRSA.
S. aureus produces a plethora of virulence factors that contribute to its success as a pathogen.
Essentially, exoproteins involved in virulence can be divided into three categories, those
involved in adhesion to host tissues, ones that interfere with immune functions of the host, and
finally those involved in tissue damage and dissemination. In general, surface-active proteins
involved in immune evasion and adhesion to host tissue are expressed early on infection and
serve to aid in establishing an infection. Surface-active proteins include fibronectin proteins,
Protein A, and coagulase. Exoproteins involved in targeting specific host cells for tissue invasion
and further immune evasion, are expressed later once infection is established and bacterial
numbers have increased significantly activating the agr quorum sensing system (Novick et al.,
1993).
The exoproteins of particular interest to this thesis are alpha-hemolysin (Hla), protein A
(Spa), and phenol soluble modulins (PSMs) as they are encoded by the core genome and their
presence are near universal in S. aureus regardless of methicillin resistance phenotype. In
addition, they have been shown to be up-regulated in the epidemic CA-MRSA strains and
demonstrated to play major roles in the pathogenesis of experimental models of pneumonia
(Diep et al., 2013; Mongkolrattanothai et al., 2003), thus, making them attractive targets for
therapeutic development against infections caused by contemporary strains.
7
Hla. S. aureus produces a number of proteins required for host immune evasion and tissue
invasion all of which play key roles in pathogenesis. Many of these exoproteins are pore-forming
that result in necrosis or lysis of a variety of host cells (Gouaux et al., 1997). Alpha-hemolysin
(Hla), one of the best-characterized exoproteins of S. aureus, is a 33kDa protein that aggregates
on host red blood cells (RBCs) and forms a heptameric barrel in the RBC membrane leading to
cell rupture (Bhakdi et al., 1996). While RBC’s are the primary target of Hla it has been reported
to have lytic capabilities on other host cells such as PMNs, epithelial cells, and platelets (Bhakdi
et al., 1988). Hla also has been demonstrated to play a role in SSTI, by stimulating keratinocyte
production of IL-1β, proliferation, apoptosis and necrosis of the basal level keratinocytes, with
the isogenic hla deletion strains causing less severe disease compared to wild-type (Prince et al.,
2012; Ragle and Bubeck Wardenburg, 2009).
Ample evidence also currently supports its role in the pathogenesis of staphylococcal
pneumonia. In a rat model of pneumonia, a Hla-producing S. aureus strain caused increased
alveolar epithelial type-1 cell damage and bronchoaveolar lavage (BAL) fluid compared to an
isogenic hla deletion mutant (McElroy et al., 1999). In another study, isogenic hla-deleted
strains exhibited less cytotoxic effects on A549 lung epithelial cells and mice infected with
isogenic hla-deleted strains were less likely to develop pneumonia and had higher rates of
survival compared to wild-type strains (Wardenburg et al., 2007). Additionally in airway
epithelial cells Hla has been associated with calcium fluxes, pro-inflammatory signaling, and
alteration of ciliary beat frequency of cells, leading to increased vascular leakage and damage
(Prince). Furthermore, vaccination against the Hla toxin conferred protection against MRSA
pneumonia in mice (Wardenburg and Schneewind, 2008). While it has been demonstrated that S.
8
aureus virulence in murine and rabbit pneumonia is attributed to Hla and CA-MRSA laboratory
strains that produce higher amounts compared to HA strains. However, little is known about Hla
production among clinical MRSA strains and if in-vitro production correlates with severity of
disease as with other bacterial infections (Li et al., 2009; Wardenburg et al., 2007).
Aside from directly damaging host cells, different studies have also demonstrated the
ability of Hla to activate the host immune system, essentially making Hla a “double edged
sword” where it can activate the immune system aiding in host defense against the pathogen, but
can also lead to an exaggerated response leading to host tissue damage. Examples of its
importance for immune response in immune activation include nucleotide-binding
oligomerization domain 2 (NOD2) activation by Hla, which leads to cutaneous defense in mice.
In this model wild-type mice mounted an immune response to Hla by stimulating secretion of
IL1β, IL-6, and lead to PMN activation, which did not occur in NOD2 mutant mice (Hruz et al.,
2009). Yarovinsky et al found that a number of cytokines, such as IL-1β, TNFα, and IFN-γ, all of
which are known to be released during S. aureus infection and protect host cells (A549 and THP-
1) from Hla damage (Yarovinsky, 2012). Recently, it was shown that Hla induces IL-17
secretion leading to a Th17 response, adding to the host humoral response (Frank et al., 2012).
Finally, loss of Hla expression in S. aureus mutants was found to correlate with increased
survival compared to wild-type S. aureus in murine abscess and wound models (Schwan et al.,
2003), suggesting that loss of Hla immune stimulation can lead to selection of these mutants
which theoretically can lead to prolonged infection during invasive infections.
On the other hand, it is known that severe infections such as sepsis and necrotizing
pneumonia are due to exaggerated host response to infection. In sepsis, over-production of
cytokines leads to peripheral blood vessel dilation resulting in low blood pressure. Acute lung
9
injury during pneumonia can occur when excess cytokines are released at the site of infection
leading to increased recruitment of PMNs, which secrete reactive oxygen species and other
compounds that aid in elimination of bacteria but also damage host cells. Bartlett et al
demonstrated that Hla plays a key role in generation of CXC chemokine gradients that lead to
increased PMN homing to the lungs during pneumonia, resulting in increased pulmonary edema
and lung injury in a murine model (Bartlett et al., 2008).
PSMs. Alpha-type phenol soluble modulins (PSMαs) are the most recently identified proteins
that contribute to virulence in experimental models of SSTI, peritonitis, pneumonia, and
bacteremia (Diep et al., 2013; Wang et al., 2007). Importantly, the role of PSMα’s in
pathogenicity has been shown in murine models of SSTI and bacteremia as well as in rabbit
models of pneumonia; the latter is considered to be the closest animal model to humans for
assessing the effect of exotoxins on neutrophils. Due to the amphiphatic nature of PSMα
peptides, they target primarily neutrophils and other cells leading to pore formation within the
cell membrane leading to inflammatory mediator release. It was also found the PSM peptides are
produced as formylated and non-formylated peptides, where a formyl group is placed on the
methionine of the peptide. The formylated PSM peptides have been demonstrated to be toxic
toward target cells compared to non-formylated (Joo et al., 2010; Wang et al., 2007).
Two types of PSMs have been identified: the α-group contains four peptides while the β
group consists of two. The α-type PSMs are potent cytolysins with PSMα3 being the most
potent, while the β-type are primarily involved in dissemination of bacteria from biofilms, and
have limited cytolytic abilities (Periasamy et al., 2012). PSMα3 is the most potent of the four α-
type PSMs and has been shown to work synergistically with other the other cytolytic virulence
10
factor PVL in causing lysis of human neutrophils (Hongo et al., 2009). Interestingly, as virulence
factors, these peptides are under unique control of the agr global regulatory system, in which the
response regulator, AgrA, primarily regulates production as opposed to RNA III (discussed
below) (Li et al., 2009).
Like Hla, PSMαs have been found to be produced in much higher amounts in CA-MRSA
strains compared to HA strains (Li et al., 2009; Yamaki et al., 2011b). Importantly,
concentrations of purified PSMα3 as low as 3.1 µg/ml have been demonstrated to cause
significant cell lysis of human neutrophils, with concentrations greater than 30µg/ml leading to
nearly 100% lysis (Hongo et al., 2009). Little is known about PSMα production of all four
peptides among clinical isolates or how their production is affected during antibiotic exposure.
Spa. Nearly all S. aureus strains produce Protein A (Spa), a 42-56 kDa surface protein encoded
by the spa gene on the core chromosome, which is present on the bacterial cell wall and is also
secreted. It is primarily involved in evasion of the host defense system during early stages of
infection (Cheung et al., 2004). Spa binds to the Fc portion of immunoglobulins, leading to
antibody inactivation and a decrease in opsonization. In in-vitro models spa deletion mutants
were more effectively phagocytosed and in murine models these deletion mutants exhibited
decreased virulence (Widaa et al., 2012).
Moreover, in recent years it has been found that aside from host immune evasion it also
plays a key role in the pathogenesis of S. aureus pneumonia through additional mechanisms
involving the inflammatory pathway and cell transmigration. Gomez et al showed Spa
recognizes TNFR1 resulting in TNFα-like responses, including activation of p38 and JNK1/2,
leading to stimulation of AP1 and NF-κB and increases in IL-8. Martin et al found that Spa can
11
induce type I INF response when in contact with airway epithelial cells (Martin et al., 2009).
Additionally, Spa was also found to induce cytoskeletal changes by activating Rho GTPases
leading to changes in actin formation, actomyocin contraction, and stress fiber formation,
resulting in tight junction dysfunction (Gomez et al., 2004; Soong et al., 2011a). Additional host
cell alteration by Spa includes the stimulation of proteases such as calpains, which lead to the
digestion of occludins and e-cadherin through the EGF receptors. These activities by Spa have
been verified in murine models of pneumonia, demonstrating not only that Spa interferes with
host immune response but also alters host cell structure allowing for bacterial transmigration
(Soong et al., 2011a).
Lastly, inflammatory responses to Spa have been shown to be dependent on the number
of short sequence repeats (SSRs) present. These SSRs are commonly used as a molecular typing
method known as Spa typing. Currently there are only 14 Spa types that have frequencies greater
than 1%, which make up 50% of reported S. aureus isolates (Garofalo et al., 2012). Garofalo et
al demonstrated that the conserved SSRs of Spa actually play a role in S. aureus virulence, where
a dose response effect in the activation of type I INF and other inflammatory cytokines can be
observed depending on the length of the region. It was found that the longer the region the
stronger inflammatory response is elicited. Most importantly, this was verified on a large
collection of clinical isolates (Garofalo et al., 2012).
PVL. The Panton Valentine-leukocidin toxin was first described in 1932 among S. aureus
strains. It was not until the 1990s that strains carrying this toxin emerged as the primary cause of
SSTI and cases of necrotizing pneumonia. PVL is synergohymenotropic bi-component toxin
encoded by two genes, luk-F and luk-S, which are located on a prophage, both components are
12
required to produce a biological effect. PVL binds to the target cell membrane and causes pore
formation at high concentrations or induce BAX-independent apoptosis at low concentrations
(Genestier et al., 2005). Lysis of host cells then results in the release of inflammatory mediators
such as cytokines, contributing to inflammation and recruitment of more host innate immune
effector cells. Neutrophils are the primary target of PVL but it can also lyse monocytes,
macrophages, and erythrocytes to a lesser extent. Recently, it has also been shown to work
synergistically with PSMα3 in the lysis of PMNs (Hongo et al., 2009).
While the exact role of this toxin in disease is inconclusive, it has been epidemiologically
linked to necrotizing fasciitis, necrotizing pneumonia, and SSTIs and thus, the presence of genes
lukF/S encoding PVL has been commonly used as a molecular marker of CA-MRSA. Initially it
was found by Labanderia et al to cause necrotizing pneumonia and severe disease in a murine
model, however after thorough investigating it was found that the experimental methodologies
were not sound and when controlled correctly by other investigators, the results from the
Labanderia lab were not reproducible (Labandeira-Rey et al., 2007; Wardenburg et al., 2007).
This was true in murine models of SSTI and pneumonia (Wang et al., 2007); however, Diep et al
recently demonstrated that in a rabbit pneumonia model, PVL seems to play a major role along
with Hla (Diep et al., 2013).
S. aureus and The Innate Immune Response.
It is likely that disease severity is not only due to exotoxins produced by S. aureus, but depends
on the interplay between MRSA virulence expression, host immune evasion, and host
inflammatory response. In a study by Daum et al, using a rat model of pneumonia and PVL (+)
MRSA, it was observed that rats which succumbed to infection had higher levels of cytokine
13
production in the first 6 hours after initial infection compared to levels of rats that did not
succumb to pneumonia, suggesting that inflammation may have a role in pneumonia
pathogenesis (Montgomery et al., 2008). Another study using a rabbit model of pneumonia found
that PVL was responsible for increased mortality, increased cytokine release, and lung
inflammation and injury through PMN-dependant mechanisms. The authors proposed a model in
which: if PVL is produced at sufficient quantities, PMNs and macrophages will be activated
leading to the release of pro-inflammatory mediators resulting in inflammation and further
recruitment of PMNs to the lung. PMNs will then enter the lung, where they will then be lysed
by the higher concentration of PVL, leading to increased damage of lung epithelium by the
spillage of PMN toxic products (Diep et al., 2010).
Evidence from the above studies suggest that damage to host cells are not only caused by
the direct effects of specific toxins towards tissues, but also contribute to an inflammatory
response that leads to further damage. A number of studies support that the inflammatory
response mediated by the innate immune system in response to bacterial products such as
lipopolysaccharide and muramyl dipeptide plays a role in the mechanisms of lung injury.
Although a number of different pathways exist in the inflammatory process, it has been
demonstrated that pathogen recognition receptors (PRR) play an integral role in acute lung injury
(Xiang and Fan, 2010). Toll-Like receptors (TLRs) for example are present in the lung, and LPS
(an activator of TLR-4), was shown to be a major cause of mortality in acute lung injury in
humans (Bernard et al., 1994).
Neutrophils (PMNs) are a major component of the innate immune system and play key
roles against bacterial pathogens such as S. aureus. PMNs are the first cells recruited to the site
of injury/infection by chemokines/cytokines (e.g. CXCL6, CXCL8) released from nearby host
14
cells as well as proteins secreted by pathogens (N-formylated peptides) (McLoughlin et al.,
2006). Once PMNs arrive at the site of infection they phagocytize invading organisms and/or
release antimicrobial granules to clear the invading organism. Essentially, the process of PMN
defense can be described in three general steps: 1) extravasation and chemotaxis to the site of
infection, 2) phagocytosis of the organism, and 3) destruction of the organism within the PMN
phagosome.
S. aureus has evolved a number of mechanisms to evade the innate immune system,
including interference with PMN functionality. As mentioned in the above section regarding
virulence factors, S. aureus strains produce exotoxins such as leukocidins and PSMα peptides
that are capable of lysing PMNs, thus decreasing the efficacy of the innate immune response.
Aside from production of exotoxins that directly target PMNs, S. aureus has developed
additional mechanisms that interfere with the functions of PMNs at all three steps listed above
and thus aid the bacterium in evading the innate immune system.
For example, formylated peptides produced by microorganisms can be sensed my PMNs
and thus serve as chemo-attractants. The chemotaxis inhibitory protein of S. aureus binds to the
formyl-peptide receptor 2 thereby blocking PMN recognition of formylated peptides such as
PSM peptides, thereby inhibiting the first step in PMN response; chemotaxis to the site of
infection (Rautenberg et al., 2011). Secondly, as mentioned in the above section of protein A,
opsonization and phagocyctosis can be inhibited by Spa, leading to yet mechanism of PMN
evasion (Forsgren and Nordstrom, 1974).
Finally, S. aureus is capable of resisting the microbicidal actions of PMNs through a
number of different mechanisms. Firstly, the “golden pigment” known as Staphyloxanthine, from
which S. aureus derives its name, is a carotenoid pigment that has been demonstrated to have
15
anti-oxidant properties that protect the bacterium from reactive oxygen species and hydrogen
peroxide (Liu et al., 2005). Secondly, enzymes such as catalase and superoxide dismutase
provide bacterial protection by eliminating reactive oxygen species produced within the
phagosome. Antimicrobial peptides (AMP) are also produced by PMNs and other host cells,
which are attracted to the negatively charged bacterial cells of both gram (+) and gram (-)
bacteria and aid in elimination. Two mechanisms are known to be employed by S. aureus to
circumvent AMP activity. The first is increased production of proteases when AMPs are
detected; which results in AMP destruction. Secondly, as AMPs are attracted to bacterial cells by
charge, S. aureus strains can increase the incorporation of D-alanine and teichoic acids that in
turn will result in increase positive charge of the membrane, thus decreasing affinity of AMPs
(Peschel et al., 2001; Peschel et al., 1999).
While S. aureus strains have a number of different mechanisms to evade the innate
immune system, the above mechanisms of PMN evasion are key to S. aureus survival from the
initial immune response. The above mechanisms taken together with the ability of strains to
produce leukocidins and other exotoxins that lyse host cells, demonstrate that virulence in S.
aureus lies at least in part by the ability to alter PMN function and/or destroy PMNs leading to
depletion and indirect host damage (Rigby and DeLeo, 2012).
Virulence Factor Regulation.
agr. Regulatory systems of staphylococcal virulence can be divided into two main groups, two-
component signal transduction systems and global transcriptional regulators. As mentioned
above, both Hla and PSMs are responsive to quorum sensing control. Their expression is
regulated by the intricate system of two component signal transduction systems, with the agr
16
(accessory gene regulator) at the center of the system, and SarA lying upstream while SaeR/S
downstream of agr (Diagram 1). The agr, is a quorum sensing system that is central to virulence
factor regulation in S.aureus. It is comprised of three promoters P1, P2, and P3. P2 directs
transcription of RNA III which encodes for the four components of the system; agrDBCA. AgrB
is a transmembrane protein responsible for processing the auto-inducing peptide of the system,
AgrD (Lina et al., 1998). AgrC is the sensor of the system and phosphorylates AgrA, upon
binding of AgrD. AgrA then binds to a region in between P2 and P3 and increases transcription
of both promoters, leading to increased transcription of all four agr components of the system by
P2 (see diagram 1) (Novick, 2003; Queck et al., 2008). P3 directs transcription of RNA III, a
regulatory RNA molecule responsible for increasing translation of Hla and also represses mRNA
translation of rot (repressor of toxin), leading to attenuated inhibition of exotoxins (Benito et al.,
2000; Boisset et al., 2007). RNAIII not only serves as a regulatory RNA molecule but also
encodes the cytotoxin γ-toxin (Hld). Recently through genome sequencing analysis, AgrA has
also been found to bind directly to a number of promoters throughout the S. aureus genome
leading to negative regulation in most cases, however of special note, the psmα promoter is under
direct positive control of AgrA.
Deletion of the agr locus has been shown to lead to a decrease in production of
exoproteins involved in virulence and pathogenesis in vitro. The critical role of agr in virulence
factor regulation was further confirmed using in-vivo mouse models of septic arthritis,
osteomyelitis, and a rabbit endocarditis model; in these studies mortality and disease severity
were decreased in agr knockout strains (Abdelnour et al., 1993; Cheung et al., 1994; Gillaspy et
al., 1995). Taken together, these observations suggest that the agr system is a key component to
virulence regulation at both the transcriptional and translational level. Furthermore, the agr
17
system has been thought to be a primary reason for the increased virulence of CA-MRSA strains
compared to HA strains since it has been observed that expression of RNA III is up-regulated in
these strains (Li et al., 2009). In addition to virulence factor regulation, agr has been shown to
affect other cellular processes and is thought to be involved in cell wall metabolism as previous
studies have found an association between vancomycin (a cell wall active agent that has been the
standard of treatment for MRSA) hetero-resistance and agr-defective strain.
SarA. The staphylcoccal accessory regulator, SarA is a 14kDa winged helix-turn-helix DNA
binding protein that is expressed early in growth (Rechtin et al., 1999). A number of promoter
regions throughout the S. aureus genome contain the SarA consensus sequence, and thus are
directly regulated by SarA which can lead to increased or decreased expression of target genes
including virulence factors (Chan and Foster, 1998; Schumacher et al., 2001). One major binding
site of SarA is the agr promoter region, where it binds as a dimer at a site between P2 and P3,
leading to full activation of the agr system at post-exponential growth phase. Hence, SarA also
regulates virulence factors through an agr-dependent manner. SarA is not only limited to
regulation of exoprotein expression but is also involved in regulation of cell wall metabolism.
Autolysins; proteins responsible for degrading the cell wall of S. aureus during cell division are
repressed by SarA, thus SarA plays a critical role in maintaining cell wall integrity (Trotonda et
al., 2009). Phenotypes of SarA knockout mutants have been noted to display reduced in vitro
virulence factor expression and to cause less severe disease in rodent animal models (Cheung et
al., 1994; Nilsson et al., 1996; Xiong et al., 2006).
18
SaeR/S. The Staphylococcal accessory element, like agr is a two-component system, and
regulates virulence through agr-dependent and independent mechanisms. SaeS provides a
method for S. aureus to sense and respond to environmental stimuli through SaeR and in many
cases leads to increase virulence factor production by binding to a conserved consensus sequence
possessed by most virulence genes, e.g. Hla (Nygaard et al., 2010; Steinhuber et al., 2003; Sun et
al., 2010). It has been shown that agr activation leads to increased transcription of Sae promoters
mediated through RNA III, however, environmental signals can override RNA III effects
(Novick and Jiang, 2003). As with SarA and agr, the SaeRS system is a critical regulatory
component for virulence factor expression and has been implicated in immune evasion as the
SaeRS system not only controls expression of adhesion proteins but also exotoxins when in the
presence of hydrogen peroxide or α-defensins, both major products of human neutrophils (Geiger
et al., 2008). While most work investigating the regulatory functions of SaeRS have been done
in-vitro, one group confirmed that the system is critical in-vivo with the use of mouse sepsis and
SSTI infection models. The investigators observed a decrease in size and intensity of wounds
caused by the CA-MRSAsaeRS knockout mutant in a cutaneous infection model as well as
decreased mortality in during bacteremia (Nygaard et al., 2010).
Current therapies and effects on virulence.
MRSA is resistant to all penicillin derivatives and first through fourth generation cephalosporins.
Currently only one fifth-generation cephalosporin antibiotic, ceftaroline, is recently approved for
the treatment of MRSA SSTIs; however its clinical effectiveness in the treatment of MRSA
invasive infections has not been studied in a prospective clinical trial. Thus, treatment of MRSA
invasive infections is currently limited to one or more of the following antibiotics: vancomycin,
19
linezolid, daptomycin, quinupristin/dalfopristin, or tigecycline. Other non-beta-lactam antibiotics
such as doxycycline, clindamycin, and sulfamethoxazole/trimethoprim are commonly prescribed
as the oral treatment options for CA-MRSA strains infections.
Since the 1960s, the gold standard of treatment has been vancomycin, a glycopeptide
antibiotic that inhibits cell wall biosynthesis by preventing cross-linking of the peptidoglycan
layer (ref). After over 50 years of clinical use, it remains the mainstay treatment of MRSA
infections. However, increasing reports of resistance and treatment failure to vancomycin,
particularly in bacteremia and pneumonia patients have questioned its clinical effectiveness.
Moreover, the Clinical and Laboratory Standards Institute redefined the breakpoint for
vancomycin susceptibility from 4 to 2 mg/L in 2006, due to the increasing reports of vancomycin
failure (CLSI 2006).
Further complicating the treatment of infections with vancomycin is the increasing report
of heteroresistant strains. By routine detection methods used in the clinical setting, MRSA strains
that exhibit heteroresistance may be considered susceptible because of the limitation in detecting
the small sub-population of less susceptible bacteria called heteroresistant vancomycin-
intermediate S. aureus (hVISA). To date, a number of studies have demonstrated a positive
correlation between vancomycin treatment failure and the presence of the hVISA phenotype in
bacteremia infections (Sakoulas et al., 2004). The hVISA phenotype is not the only factor, which
plays a role in the possible treatment failures with vancomcyin. The pharmacokinetic and
pharmacodynamic properties also play a role in treatment failures. Vancomycin requires
concentrations of 4-5 times the strains minimum inhibitory concentration for optimal therapeutic
effects, meaning adequate penetration of vancomycin to the site of infection must be achieved.
20
However, with vancomycin being a large molecule, it has been shown to have poor penetration
to solid organs for example, lungs, and also has poor penetration into the cerebrospinal fluid.
Among the alternative agents for vancomycin; linezolid, daptomycin, and tigecycline are
the most commonly prescribed within the clinical setting. Daptomycin is a lipopeptide that acts
on the bacterial cell wall by binding to the cell membrane causing rapid depolarization of
membrane potential eventually leading to cell death. Tigecycline and linezolid on the other hand
act through inhibiting protein synthesis inhibitors that inhibit translation via binding to
components of the bacterial ribosome. Being structurally similar to tetracyclines, tigecycline
works by binding to the 30S ribosomal subunit and prevents tRNA binding to the ribosome. The
oxazolidinone, linezolid, inhibits protein synthesis via binding to the 23S rRNA of the 50S
ribosomal subunit, inhibiting the formation of the initiation complex (70S ribosome), thus
inhibiting protein translation. Historically, the gold standard protein synthesis inhibitor for
MRSA infections has been clindamycin, although it is not commonly used to treat serious
infections, it is routinely given in the outpatient settings for non-complicated SSTIs.
Clindamycin inhibits protein synthesis via binding to the 50S subunit leading to inhibition of
ribosomal translocation.
Considering the impressive arsenal of virulence factors mentioned above that contribute
to the success of MRSA as a pathogen, previous studies have investigated the effects of protein
synthesis inhibitors on exoprotein expression. Specifically, clindamycin has been shown to
decrease secreted proteins such as Hla, δ-hemolysin, coagulase and TSST-1 (Blickwede et al.,
2005; Doss et al., 1993; Herbert et al., 2001; Schlievert and Kelly, 1984; vanLangevelde et al.,
1997). Recently, sub-inhibitory concentrations (½, ¼, and ⅛ MIC) of linezolid and clindamycin
have also been shown to markedly decrease PVL production (Dumitrescu et al., 2008;
21
Dumitrescu et al., 2007). While these results were from in-vitro experiments, similar inhibitory
effects with linezolid and clindamycin have also been observed in studies using laboratory
reference strains or few clinical isolates in cell culture models as well as animal models of
infection (Koszczol et al., 2006; Yanagihara et al., 2008). Since the discovery of PSMα peptides,
none had investigated the effects of antibiotics on their production. Considering the importance
of PSMs contribution to virulence and their unique regulation, we sought to determine the effects
of protein synthesis inhibiting antibiotics on PSM production. Notably, we have tested a large
collection of clinical isolates for baseline PSMα
1-4
peptide production to confirm previous
observations using laboratory reference strains and few clinical strains by others.
In addition to exoprotein inhibitory effects, one study demonstrated that proteins involved
in metabolism and stress response were induced by sub-inhibitory concentrations of linezolid
when a complete analysis of all exoproteins using 1D and 2D electrophoresis was performed
(Bernardo et al., 2004). This demonstrated that not all protein production is inhibited by this
class of antibiotics and inhibitory effects primarily affect proteins required for virulence.
Additionally, it has been shown that sub-inhibitory concentrations of β-lactam antibiotics can
also lead to up-regulation of virulence toxins secretion, particularly nafcillin and oxacillin. This
was demonstrated with PVL and Hla production, which increased in the presence of nafcillin
(Dumitrescu et al., 2008; Kernodle et al., 1995; Stevens et al., 2007b). These data suggest that β-
lactams can induce virulence factor expression and whether protein synthesis inhibitors can have
these effects deserves further investigation.
Induction of exotoxin production from antibiotic exposure has potential implications on
the empiric choice and dosing of antibiotics for treatment of S. aureus infections and resultant
outcomes. The general assumption made by clinicians that protein synthesis inhibitors have a
22
beneficial effect even at concentrations below the minimum inhibitory concentration needs
validation before this notion can be applied to manage patients in the clinical setting. This notion
evolved from previous in-vitro studies on S. aureus enterotoxins, PVL, and Hla as well as
clinical practice in the treatment of Streptococcus pyogenes necrotizing fasciitis, where protein
synthesis inhibitors are commonly given as adjunct therapy. As the focus in new treatments is
shifting towards non-traditional approaches to inhibit toxin production due to increased
emergence of strains resistant to standard antibiotic therapy, it is imperative to have a complete
understanding of the clinical and molecular epidemiology of S. aureus infections and the
regulatory systems that govern virulence and their response to antibiotic exposure.
Summary.
Methicillin resistant S. aureus is now recognized by the Infectious Disease Society of
America as one of six bacteria that pose an immediate threat to public health in the United States.
Once restricted to the healthcare settings, MRSA has now emerged as a leading cause of
infections within the community settings. While rare, severe infections caused by this organism
have been reported, and it is thought that exoproteins expressed by these organisms in part
contributed to the severity of these infections. Numerous studies suggest that the addition of
protein synthesis inhibiting antibiotics may be beneficial in treatment of MRSA infection due to
inhibition of exoproteins. Furthermore, there have been increasing reports of decreased
susceptibility to commonly prescribed antibiotics. Thus, the goal of this thesis was to 1) address
if the changing epidemiology of MRSA infections reported elsewhere is occurring within our
locale, 2) identify clinical or microbiologic characteristics that can help guide antibiotic selection
against MRSA infections, 3) evaluate the contribution of exotoxins and resistance to patient
23
outcomes, and finally 4) investigate the anti-virulence potential of antibiotics prescribed for the
treatment of MRSA infections.
Diagram 1. Regulation of the agr system
Adapted from Bronner et al. FEMS Microbiology Reviews 28 (2004) 183–200
SarA
Sae
R/S
+
24
Chapter 2.
Clinical and molecular epidemiology of MRSA infections and associated outcomes
Bacteremia is the leading invasive infection caused by MRSA, however uncertainty remains
about the contribution of CA-MRSA strains in bacteremia. Some studies have found a high
percentage of isolates belonging to CA-MRSA strains ranging from 37% - 56% (Chua et al.,
2008; Fridkin et al., 2005; Maree et al., 2007), while other groups have reported smaller numbers
as low as 2% (Fridkin et al., 2005; Klevens et al., 2007). These studies have demonstrated a shift
in the epidemiology of MRSA bacteremia in other parts of the country but few studies have been
conducted within the Los Angeles area.
Pneumonia is the next most common invasive infection following bacteremia. Depending
on the place of acquisition and healthcare exposure, pneumonia can be categorized as CAP,
HCAP, HAP, and VAP. MRSA has now become a significant cause of pneumonia in all four
categories. It alone in 2002-2003, was the etiology of 16% of pneumonia cases that presented to
59 US hospitals; 9% of which were defined as CAP. In the healthcare setting, MRSA accounts
for 20-40% of HAP and VAP cases (Rubinstein et al., 2008). Moreover, a surveillance study of
severe CA-MRSA infections in Georgia from 2005 to 2007, 6% of patients had MRSA
pneumonia and 29% of those patients died (David and Daum, 2010). Likewise, a recent article
investigating HAP caused by S. aureus in 235 institutions in 38 different countries found only
8% of S. aureus strains were CA (Sharma-Kuinkel et al., 2012). Furthermore, severe necrotizing
pneumonia has been reported by CA-MRSA strains carrying the PVL genes in a small case study
(Gillet et al., 2002), suggesting that PVL may be responsible for the severity of this disease.
Currently there are two methods used to distinguish between CA vs HA-MRSA. One
method is the CDC clinical classification system that is based on the timing of isolation and
25
previous healthcare exposure. However, the major limitation of this system is that actual CA-
MRSA genetic background lineages could be misclassified, thus their true contribution may be
skewed. Hence, molecular markers are increasingly being used to categorize MRSA strains in
order to truly identify the contribution of CA vs HA strains in disease. A number of different
molecular techniques can be employed to identify strain background; pulse field gel
electrophoresis being the gold standard. However, the most commonly used technique is PCR for
detecting SCCmec type and PVL genes, due to its simplicity and short turnaround time. CA-
MRSA strains are distinguishable from HA-MRSA by the SCCmec element, with SCCmec IV
and V predominating in CA-MRSA strains while SCCmec II and III in the latter strains (Huang
et al., 2006). In addition, 75-90% of CA-MRSA strains harbor the genes lukF-S genes encoding
the PVL toxin compared to less than 5% of HA-MRSA strains (Naimi et al., 2003; Vandenesch
et al., 2003). More importantly, HA-MRSA strains are characterized by multidrug-resistant
phenotype including recent emergence of strains with reduced susceptibility to vancomycin
(MIC > 1 µg/ml) that are associated with treatment failure (Hidayat et al., 2006; Musta et al.,
2009).
Considering the CDC definitions and distinguishing features between community and
healthcare-associated MRSA, clinicians may presume patients being admitted from the
community to be infected with more vancomycin-susceptible MRSA strains and thus prescribe
less aggressive dosing of vancomycin empirically or have a lower tendency to consider
alternative agents based on susceptibility testing results. Recent epidemiologic evidence
indicates that CA-MRSA strains are becoming endemic within hospital settings and are likely to
undergo the same selection pressure as HA-MRSA strains for reduced susceptibility to
vancomycin (Maree et al., 2007; Seybold et al., 2006).
26
We hypothesized that CA-MRSA strains would account for a significant portion of
infections among hospitalized patients and that markers from bacteria or host exist to guide
therapy. Thus, we sought to determine the contribution of CA-MRSA and HA-MRSA strains as
determined by molecular markers at our institution in three of the most common types of
infection; skin soft tissue infection, pneumonia, and bacteremia, with an emphasis on the
invasive diseases. Additionally, with the clinical information collected, our goal was to identify
patient characteristics and clinical features that may predispose to infection caused by HA-
MRSA vs CA-MRSA and identify patient and bacterial markers that can be utilized for treatment
guidance.
Methods.
Study Design and Population. The following retrospective cohort studies were performed at
Huntington Hospital, a 625-bed community teaching hospital located in Pasadena, California.
The institutional review board approved the studies. Patients hospitalized between July 2005 and
June 2011 were included if all of the following were met: age > 17 years, had an infection as
defined by CDC criteria (Horan et al., 2008) that is caused by MRSA, bacterial isolate and
medical charts were available. Medical charts of patients were reviewed and relevant
demographic, laboratory, and clinical data were recorded onto a structured data collection form.
Specifically, the following was obtained from the medical charts: age, gender, place of residence
prior to admission, any healthcare exposure as defined by CDC during the prior year, onset of
infection within or beyond 48 h of hospitalization, comorbid conditions, underlying severity of
illness as measured by APACHE II score (calculated based on the worst values within 24 h of
hospital admission), type of infection, vancomycin trough values, duration of ICU or hospital
27
stay, clinical laboratory values, and antibiotic therapy. All data were subsequently entered into a
relational database management program (Microsoft Access) for data analysis.
Study Definitions. Pneumonia, necrotizing fasciitis, bloodstream infections, and all other sterile
site infections were defined as invasive infections. HA infections were defined as those which
MRSA was identified > 48 hours after admission, in patients with a history of hospitalization,
surgery, dialysis, permanent indwelling catheter or percutaneous medical device, or known prior
positive culture for MRSA, or resided in a long-term care facility within 1 year of the MRSA
culture date; otherwise they were considered CA per CDC definitions.
Patients were included in the study for community-onset pneumonia if they met the
above population criteria and the following: 1) diagnosis of pneumonia within 48 h of hospital
admission, 2) respiratory culture obtained during admission, 3) meeting CDC criteria for
pneumonia, 4) received at least 48h of effective therapy and 4) age >17 years. For patients with a
positive respiratory culture, only those with single bacterial pathogen isolated were included. For
specific inclusion in the bacteremia study, patients needed to meet the following additional
criteria: 1) growth of MRSA from blood culture and 2) CDC criteria for bloodstream infection.
Persistent bacteremia was defined as 7 days or greater positive MRSA blood cultures despite
receiving anti-MRSA therapy.
Clinical responses of all patients throughout treatment were evaluated at 72h and end of
treatment. A complete response was defined as resolution of fever, leukocytosis, local signs of
infection, return of abnormal vital signs and altered mental status to baseline; a partial response
was improvement of these conditions without complete resolution. An early response measured
at 72h was defined as meeting the following: defervescence (temperature ≤ 37.6 C), return of
28
white blood cell count with differential towards normal range, and subjective feeling of well
being. Failure was defined as no improvement or worsening of signs and symptoms of infection
at 72h and at end of treatment, which included relapse and in-hospital death. Relapse was
defined as recurrence of infection within 60 days after discontinuation of therapy. Death was
attributed to infection if it occurred in the context of septic shock or endocarditis or if the patient
had signs of infection and persistent positive blood cultures with no other plausible explanation
for death.
Microbiologic Testing. Non-duplicate MRSA isolates were collected from infected patients and
stored in cryovials at -80°C for later testing. Isolates were identified as S. aureus by the tube
coagulase method. Oxacillin resistance was confirmed using oxacillin agar screen test with 6
mg/L oxacillin on Mueller-Hinton agar containing 2% NaCl according to CLSI guidelines.
Minimum inhibitory concentrations (MICs) of vancomycin, were determined by Etest method
according to manufacturer’s instructions (BioMerieux, Durham, NC). Briefly, strains were
grown overnight on blood agar plate prior to inoculating Mueller Hinton agar with bacterial
suspension at 0.5 McFarland standard. Each plate was incubated for 24 hours at 37°C after
placement of Etest strip for the respective agent. MIC was read at the concentration where the
eclipse of growth inhibition intersects the strip; where colonies occurred within the inhibition
eclipse, the higher value was taken as the MIC.
Polymerase Chain Reaction (PCR) and Evigene Assays. The Evigene-PVL assay, a nucleic
acid hybridization test (Evigene-PVL, AdvanDx, Woburn, MA) was performed to screen all
strains to detect the presence of lukF-S genes encoding the Panton-Valentine leukocidin (PVL)
29
toxin according to the manufacturer’s instructions. The assay was selected because of its
potential utility at the bedside: ease of use, rapid turnaround time of 3 hours for results, and high
sensitivity and specificity compared to results from PCR-based assay. An MRSA isolate was
considered PVL-positive if the well was red in color with an optical density (OD)
492
>0.325 and
PVL-negative if the well was clear or pink in color with an OD
492
≤0.325. Ambiguous results
were confirmed by PCR using previously published primers and conditions (Bonnstetter et al.,
2007).
NRS384 and NRS22 strains served as positive and negative controls, respectively.
SCCmec typing by PCR was performed using previously published primers and conditions to
better define the molecular epidemiology of these strains (Boye et al., 2007) (Table 1). The
master mix consisted of 1.0 µmol of each primer, 0.4mMdNTP’s, PCR buffer 10X, PCR
enhancer 10X, 2.5U Taq polymerase, and 3mM MgCl
2
. PCR conditions were as follows: 95°C
for 5 minutes, 29 cycles of: 95°C for 1 minute, 55°C for 1 minute, 72°C for 1 minute, and final
extension at 72°C for 10 minutes.
Table 1. PCR primers for SCCmec typing and detection of PVL.
Targets
Forward/Reverse
Primers
SCCmec
typing
ccrA2-‐B
β:
5’-‐ATTGCCTTGATAATAGCCYTCT-‐3’
α3:
5’-‐TAAAGGCATCAATGCACAAACACT-‐3’
ccrC
ccrCF:
5’-‐CGTCTATTACAAGATGTTAAGGATAAT-‐3’
ccrCR:
5’-‐CCTTTATAGACTGGATTATTCAAAATAT-‐3’
IS1272
1272F1:
5’-‐GCCACTCATAACATATGGAA-‐3’
1272R1:
5’-‐CATCCGAGTGAAACCCAAA-‐3’
mecA–IS431
5RmecA:
5’-‐TATACCAAACCCGACAACTAC-‐3’
5R431:
5’-‐CGGCTACAGTGATAACATCC-‐3’
lukS-‐F
5
luk-‐PV-‐1:
5’-‐ATCATTAGGTAAAATGTCTGGACATGATCCA-‐3’
luk-‐PV-‐2:
5’-‐GCATCAACTGTATTGGATAGCAAAAGC-‐3’
30
Statistical Analysis. Continuous variables were compared by student t-test or Wilcoxan Rank
Sum test. Dichotomous data was compared using Chi square with Yates correction or Fisher’s
exact test where appropriate. All statistical analyses were performed using GraphPad Prism
(version 5.0). A p-value of ≤ 0.05 denotes statistical significance.
For analysis of vancomycin susceptibility data, study groups were compared on the
following to differentiate those infected with high vancomycin MIC strains: demographics, HA
or CA, residence in home vs healthcare facility prior to admission and at onset of infection,
invasiveness of infection, presence of PVL-encoding genes, and SCCmec type of the infecting
strains.
When analyzing data of clinical response of bacteremia patients to vancomycin, patients
were grouped based on response to vancomycin assessed on day 3 to compare baseline host and
microbiologic characteristics of the infecting strains and to analyze the impact of early treatment
response on eventual outcomes. Subgroup analysis was also performed to determine the
relationship between final treatment received and eventual outcomes by stratifying the patients
based on receipt of vancomycin for entire treatment course or switched to alternative agent(s)
during the course of infection. Confounding factors were modeled in a multivariate regression
analysis to identify independent predictors of eventual treatment failure which included: age,
APACHE II score, ICU admission, vancomycin MIC >1 µg/ml, unbound vancomycin trough
concentration below 4-5x MIC, vancomycin therapy from initiation to end, and 72 h non-
response.
31
Results.
Clinical, epidemiological, and microbiologic variables of patients infected with MRSA
strains from 2005 – 2007.
Patient characteristics. A total of 186 adults hospitalized for any type of MRSA infection were
evaluated from 2005-2007 (Yamaki et al., 2011a). The study cohort was represented by an
elderly population with a median age of 70.5 years (IQR: 55, 82). Nearly half of the patients
(46%, 86/186) were admitted from home. A total of 129 patients (72%) had diagnosis of MRSA
infection within 48 hours of hospital admission; of those, only 49% (n=63) were classified as
having community-associated infections per CDC definition. Cardiovascular disease (57%,
107/186) was the leading comorbid condition followed by diabetes (34%, 63/186) and renal
disease (18%, 34/186). Only 11% (21/186) reported a history of MRSA infection within 6
months of admission. Most patients (67%, 125/186) were admitted for invasive infections; the
remaining patients (33%, 61/186) had skin and soft tissue infections. The overall median
APACHE II score was 11 (IQR: 5, 18), with 12% of patients (23/186) requiring admission to the
intensive care unit (Table 2).
32
Table. 2 Demographics, characteristics, and spectrum of disease of patients infected with
MRSA strains from the 2005 – 2007 cohort.
Characteristics PVL(+)
n=82 (%)
PVL(-)
n=104 (%)
p-value
Demographics
Age (yrs, median, IQR)
Female
APACHE II (median, IQR)
% Above IBW (median, IQR)
Co-morbidities
Diabetes
Cardio Vascular Disease
Renal Disease
Cerebral Vascular Accident
Other
1
Spectrum of Disease
SSTI (%)
Invasive Infections
Pneumonia
Bacteremia
SSTI à Bacteremia
59 [46,76]
45 (51)
2 [0,5]
28.53 [15, 61.6]
22/82 (27)
44/82 (54)
2/82 (2)
5/82 (6)
13/82 (16)
53 (65)
32 (39)
16/82 (20)
10/82 (12)
3/50 (6)
76 [62,83]
40 (40)
13 [8.25,22]
8.01 (-6.9, 28.95)
36/104 (35)
69/104 (66)
8/104 (8)
22/104 (21)
33/104 (32)
19 (18)
93 (89)
61/104 (59)
22/104 (21)
8/11 (73)
<0.0001
0.18
<0.0001
0.0008
0.49
0.0785
0.189
0.0056
0.0203
<0.0001
<0.0001
<0.0001
0.108
<0.0001
NOTE: PVL was used as a marker of CA-MRSA. The table demonstrates differences in patient
characteristics and spectrum of disease between CA vs HA-MRSA strains.
Community vs hospital-associated MRSA. Determination of CA vs HA-MRSA was performed
using the two methods described above, the clinical CDC criteria and by PCR for CA-MRSA
epidemiologic markers (PVL and SCCmec type). Based on the CDC criteria 34% (64/186) of
patients were classified as CA-MRSA with the remaining classified as HA. A total of 84 (45%)
isolates were molecularly classified as CA-MRSA based on SCCmec and PVL typing.
Interestingly when comparing clinical vs molecular classifications, we found that nearly 1/3 of
our patients (31%) were classified as HA indicating that they had previous healthcare exposure
33
but had been infected with CA-MRSA strains. Strains classified as CA-MRSA by clinical criteria
were found to be comprised of 81% CA-MRSA by molecular marker (Figure 1).
Figure 1. Distribution of PVL (+) MRSA strains among patient meeting clinical criteria for
CA or HA-MRSA as defined by CDC
CA-MRSA strains (by use of PVL marker) were responsible for the majority of CA infections as expected
(left), however these strains were responsible for nearly 1/3 of HA infections which suggests a shift in
epidemiology by CA-MRSA strains.
Spectrum of disease. As mentioned, MRSA is capable of causing a wide spectrum of diseases. In
our cohort of patients, 33% of infections were non-invasive skin soft tissue infections (SSTIs).
The remaining 125 cases were invasive infections, pneumonia accounted for 62% of these cases,
bacteremia 34%, and others were necrotizing fasciitis and UTI. CA-MRSA strains (as classified
by molecular markers) caused more SSTIs in comparison to HA-MRSA strains (65% vs 18%, p
< 0.0001). Interestingly, HA-MRSA strains were also more likely to cause pneumonia compared
to CA (59% vs 20%, p < 0.0001), but were not significantly more likely to cause bacteremia
(Table 2).
81%
31%
69%
34
Molecular epidemiology in pneumonia and bacteremia patients.
As mentioned earlier, there have been a number of reports describing the changing epidemiology
and contribution of CA-MRSA strains in the healthcare settings. In bacteremia, CA-MRSA
strains have been found to contribute significantly, in some cases more than one-third of strains
have been reported as CA-MRSA (Chua et al., 2008), while others reported smaller numbers.
The contribution of CA-MRSA in pneumonia is less well studies as in bacteremia (Sharma-
Kuinkel et al., 2012). Hence, we sought to investigate the contribution of CA-MRSA strains in
both bacteremia and pneumonia.
Pneumonia. Of the 134 unique patient isolates analyzed, 69 (52%) strains were SCCmec IV
while 59 (44%) were SCCmec II; 91% and 1% of those strains carried the PVL genes,
respectively. Only one SCCmec V strain was identified, which was PVL (-), and five isolates
were non-typable by the above methods and were PVL (-). No SCCmec I or III were found in
our study population. Surprisingly, using the CDC clinical definitions of CA-MRSA, only 5
isolates (4%) of isolates were defined as CA-MRSA, all 5 of these isolates we SCCmec IV and
PVL (+). This suggests that most of our patient population who are admitted with MRSA
pneumonia have some type of MRSA history or previous healthcare exposure.
Bacteremia. Of the 136 bacteremia isolates tested for SCCmec type, 69 (51%) of these isolates
were considered CA-MRSA and of these, 67 isolates were SCCmec IV and the remaining were
SCCmec V. 84% of CA-MRSA strains carried the genes encoding PVL, while only 6% of the
HA-MRSA isolates carried these genes. Of the 67 HA-MRSA strains, SCCmec II was the most
common cassette detected with 64 isolates being SCCmec II, the remaining three isolates were
35
SCCmec III. We found that the majority (56%) of bacteremia cases that were identified at 48
hours or less were caused by SCCmec IV/V strains, compared to 42% of infections diagnosed
after 48 hours, however only 18% of CA-MRSA infections would meet the CDC clinical
definition.
Patient characteristics and outcomes. In an effort to identify risk factors that may play a role in
BSI caused by different strains of MRSA we examined characteristics of patients infected with
CA-MRSA and HA-MRSA as defined by SCCmec type. The majority of the patients was either
residents of skilled nursing facility or had been hospitalized within 3 months (75%).
Cardiovascular disease was the leading comorbid condition (70%), followed by diabetes mellitus
(47%) and renal disease (46%). In 30% of all BSI cases the source of the infection was unknown.
When the source of infection was known, line infections and prosthesis were the most common
causes of BSI. In comparing sources of infection between HA-MRSA and CA-MRSA, line
infections were the most common source among HA strains while wound infections leading to
BSI was the leading cause in CA strains. Line or prosthesis were the most frequent causes of
infection regardless of timing of infection (Figure 2). History of MRSA infection and
vancomycin usage did not differ between patients infected with CA vs HA-MRSA.
Figure 2. Distribution of source of MRSA bacteremia (n=136).
Unknown source was the most frequent source of bacteremia, followed by catheter line and prosthetic
material.
18
10
19
8
5
30
2
8
prosthesis
undrained
abscess
line
pna
other
unknown
post-‐op
wound
36
Consistent with published literature, we also found patients infected with CA-MRSA
strains also tended to be younger and more overweight (p = 0.02). As expected we also found
that patients infected by strains with SCCmec types associated with CA-MRSA were more
frequently acquired from home (OR = 2.3, CI [1.159 to 4.582]) and HA-MRSA associated
SCCmec types from skilled nursing facilities (OR 2.46, CI [1.230 to 4.909], p = 0.0102). No
differences between the two groups were observed in regards to recent prior hospitalizations as
well as APACHE II scores upon admission, indicating that patients had similar severity of
underlying disease. Additionally, co-morbid conditions such as cardiovascular disease, diabetes,
and renal disease were equally common amongst both patient groups (Table 3).
Table 3. Patient characteristics and outcomes in the bacteremia cohort.
Characteristics
SCCmec
IV,
V
(N=69)
SCCmec
II,
III
(N=67)
Age,
yrs
(mean
+
SD)
64.4
+
18.4
70
+
14.4
Male/Female
18/37
(33%/67%)
21/29
(42%/58%)
Weight,
kg
(mean
+
SD)*
82
+
22
72
+
17
Place
of
residence:
SNF**
Hospitalization
<
3mo
Home**
24
(35%)
20
(29%)
42
(61%)
38
(57%)
20
(30%)
20
(30%)
APACHE
II
a
(mean
+
SD)
16
+
7.4
17.6
+
6.2
ICU
Admission
24
(35%)
20
(30%)
History
of
MRSA
27
(39%)
22
(33%)
Recent
VAN
therapy
b
20
(29%)
15
(22%)
Hospital
LOS,
d
a
(median,
IQR)
ICU
LOS,
d
(median,
IQR)
17
(10-‐30)
8
(1-‐15)
15
(8-‐25)
5
(2-‐8)
Time
to
achieve
clinical
stability,
days
(median,
IQR)
4
(3-‐7)
3
(1-‐5)
72hr
Response
46
(67%)
41
(61%)
37
End
of
treatment
Outcome:
Clinical
Response
Failure
46
(67%)
23
(33%)
44
(66%)
23
(34%)
Endocarditis
8
(12%)
10
(15%)
Mortality,
overall
8
(12%)
10
(15%)
hVISA
9
(13%)
8
(12%)
NOTE:
a
=
VAN
alone
or
in
combination,
b
=
Less
than
60
days
ago,
*
p
<
0.05,
**
p
<0.001
Upon admission only 80% of patients infected with CA-MRSA were started with anti-
MRSA therapy within 24 hours of admission, as opposed to 91% of patients infected with HA
strains. By 48 hours however, nearly all patients (>96%) received anti-MRSA therapy.
Vancomycin was the primary antibiotic selected for treatment in both groups, with 88% of
patients receiving it. Essentially, no significant differences were observed between patients
infected with MRSA strains that were CA or HA as determined by SCCmec type. Approximately
two-thirds of patients in each group responded by 72 hours of anti-MRSA treatment, although
patients infected with HA strains reached stability faster compared to those infected with CA
strains (p = 0.018). We found that ~20% of patients in our population had persistent bacteremia,
with no differences observed between the CA and HA groups. Overall, end of treatment
outcomes were similar in both CA and HA groups with the same percentage (66%) of patients
responding at 72 hours. No differences were observed in mortality which was 13% and 11% for
HA and CA, respectively.
Interestingly, for patients with pneumonia when grouped according to CA vs HA as in
the bacteremia study we found no differences in patient characteristics or outcomes. No
significant differences existed between the number of co-morbidities, patient weight, age, or sex.
The severity of pneumonia as assessed by the pneumonia severity index (PSI) score or PSI class
of pneumonia did not differ between groups, nor did the number of failures, readmissions,
38
APACHE II score, or deaths. Only place of residence prior to admission and administration of
vancomycin was similar to that of the bacteremia cohort. 73% of the pneumonia cohort came
from skilled nursing facilities, and patients were evenly distributed between CA and HA-MRSA.
85% of pneumonia patients received vancomycin therapy for treatment, nearly the same
percentage as those in the bacteremia cohort.
Patient and bacterial markers that can be utilized for treatment guidance.
Vancomycin has been the drug of choice for MRSA infection since its introduction in the
late 1950’s. In most of our cohorts above, 85% had received vancomycin as empiric and directed
therapy. Unfortunately, with the increase in MRSA infections and increase usage of vancomycin,
there have been a number of reports of treatment failure with vancomycin. Persistence of
infection is commonly described in bacteremia patients who receive proper anti-MRSA therapy
yet fail to clear the bacteria from the blood, with vancomycin being the most common drug
therapy associated with persistence. Here we identify a timeframe for vancomycin therapy
assessment, which can be used to determine if treatment is adequate based on response and can
provide a window of opportunity to change to an alternative therapy if response if suboptimal.
Early response in bacteremia with vancomycin. Treatment outcomes from patients with MRSA
bacteremia were analyzed at 48h and 72h after the initiation of vancomycin treatment. Out of the
136 patients with MRSA bacteremia, 111 were initiated on vancomycin therapy for 48h or
longer. Using this group of patients we analyzed patient outcomes to identify independent factors
that contribute to eventual treatment failure.
Clinical response was assessed at two time points to determine early response on day 3
39
after initiation of vancomycin therapy and later at the end of therapy. When exploring factors
that may have contributed to treatment failure we found no significant association with type of
MRSA strains (CA vs HA), susceptibility to vancomycin, patient characteristics including age,
history of MRSA infection, source of infection or co-morbid conditions. Of the 111 patients
evaluated, 62% of patients (69/111) initiated on vancomycin therapy demonstrated early
response at 72h. When patients were compared based on initial response, those patients that were
early responders were significantly less likely to eventually fail treatment (19%, 13/69 vs. 57%,
24/42; p<0.0001) and die from infection (1%, 1/69 vs 29% 12/42, p<0.0001). In addition, these
patients were less likely to have persistent bacteremia (17%, 12/69 vs 29%, 12/42, p = 0.23), had
shorter overall length of stay (median 13 d vs 16.5 d, p = 0.11) and ICU stay (3 d vs 8 d, p =
0.19), however statistical significance was not reached in these cases.
Treatment outcome was also analyzed based on whether patients received vancomycin
therapy throughout the entire treatment course or switched to alternative agents during the course
of treatment. The rates of end of treatment failure with patients who had persistent bacteremia
and death did not differ between the two groups. Although, when comparing the subset of
patients who did not achieve early response at 72h after initiation of vancomycin therapy, those
who continued on vancomycin until the end of treatment (n=14) showed a trend towards worse
outcomes when compared to the cohort who were switched to alternative agents (n=21): end of
treatment failure, 64% vs 43%; persistent bacteremia, 38% vs 24%, and death, 38% vs 10%. The
differences observed did not achieve statistical significance likely due to small sample size.
We further tested our findings in a multivariate logistic regression analysis controlling for
a number of potential confounders which included age, APACHE II score, vancomycin unbound
trough values, and vancomycin MIC values >1 mg/L. Non-response at 72h remained the
40
strongest predictor for end of treatment failure (OR 24.9, 95% CI: 4.80-129.22; p<0.0001).
Parameters differentiating patients infected with high vs low vancomycin MIC strains.
Vancomycin is currently the gold standard of treatment for MRSA infections, and is commonly
given as empiric therapy to patients who are admitted to the hospital with suspected MRSA
infections. The activity of vancomycin has been shown to be time-dependent, meaning that drug
concentrations should be maintained above the bacterial MIC at all times. Concentrations of 10x
the MIC are required for optimal exposure, thus if an isolate has an MIC of 1 µg/ml then
vancomycin serum concentrations should be 10 µg/ml. To complicate matters, vancomycin is
associated with a number of different adverse reactions, including nephrotoxicity. Nephrotoxicity
has been associated with higher concentrations of vancomycin (>15 µg/ml) and patients with
concomitant nephrotoxic agents or those who have renal insufficiencies are at increased risk for
nephrotoxicity (Hidayat et al., 2006). Meaning that patients with MRSA strains that have MIC
values of 1.5 µg/ml or greater, may be at risk for nephrotoxicity if treated with high dose
vancomycin. Hence, we sought to determine if molecular markers for CA-MRSA can predict
low vs high vancomycin MIC, in an effort to guide empiric vancomycin dosing to minimize
unnecessary risk for nephrotoxicity with high dose therapy.
Patient data and vancomycin susceptibility from 180 patient isolates during the time
period from 2005-2007 in all types of MRSA infections were used to determine if patient or
bacterial parameters can guide empiric vancomycin dosing. Vancomycin usage as empiric and
directed therapy was determined for the study group as well as demographics of patients based
by vancomycin MIC. Using the clinical and molecular data from the 2005-2007 cohort of
patients presented in the section above, we analyzed vancomycin susceptibility to determine if a
41
correlation exists between CA vs HA-MRSA infections and if differences existed depending on
the definition used. Vancomycin susceptibility was grouped as MIC ≤ 1µg/ml or > 1 µg/ml.
Vancomycin was prescribed in 94% of patients who received empiric antibiotic therapy
with activity against MRSA. About one-third (32%, 57/180) of infected strains had high
vancomycin MIC (>1.0 µg/ml). With respect to demographics and clinical characteristics, those
infected with high MIC strains (n = 57) did not differ in age (71 vs 69 y), APACHE II score
upon admission (12 vs 11), history of MRSA infection within 6 months (11% vs 12%), and
residence prior to admission (home, 46%; long term care, 54%; or acute care facility, 52%)
compared to those infected with low MIC strains. Most invasive cases were pneumonia (61%,
73/119) and/or bloodstream infection (27%, 32/119); one strain caused necrotizing fasciitis. A
slightly higher proportion of patients with invasive infections were infected with high
vancomycin MIC strains compared to those with skin and soft tissue infections (33%, 40/119 vs
28%, 17/61; p > 0.05) (Table 4).
42
Table 4. Demographics, patient characteristics, and disease spectrum of strains with low or
high vancomycin MIC.
Characteristics
VAN
<=1,
n=123
(%)
VAN
>1,
n=57
(%)
Demographics
Age
(yrs,
median,
IQR)
Female
APACHE
II
(median,
IQR)
MRSA
History
ICU
admission
due
to
MRSA
Residence
Prior
to
Admission
Home
SNF
Hospital
Homeless
Co-‐morbidities
Diabetes
Cardio
Vascular
Disease
1
Renal
Insufficiency
Dialysis
Cerebral
Vascular
Accident
Pulmonary
Disease
2
Malignancy
Immunosuppressed
Liver
Disease
Spectrum
of
Disease
SSTI
Invasive
Infections
Pneumonia
Bacteremia
Other
69
[51,81]
56
(46)
11
[5,17]
15
(12)
12
(10)
56/123
(46)
60/123
(49)
4/123
(3)
3/123
(2)
13/123
(11)
*
74/123
(60)
8/123
(7)
13/123
(11)
21/123
(17)
22/123
(18)
13/123
(11)
5/123
(4)
7/123
(7)
50
(41)
79
(64)
49/123
(40)
20/123
(16)
2/123
(2)
71
[55,82]
29
(51)
12
[5,19.5]
6
(11)
9
(16)
26/57
(46)
29/57
(50)
2/57
(4)
0/57
(0)
23/57
(40)
*
33/57
(58)
3/57
(5)
10/57
(18)
11/57
(19)
11/57
(19)
4/57
(7)
0/57
(0)
1/57
(2)
17
(35)
40
(70)
24/57
(42)
12/57
(21)
4/57
*p value < 0.0001
;
1 –
CHF, Afib, HTN,
2
- Asthma, COPD, Emphysema,
When patients were grouped based on healthcare exposure risk per CDC definition, the
proportion of strains with high vancomycin MIC was similar between CA-MRSA and HA-
MRSA groups, 27% (13/47) vs 33% (44/133), respectively. Those who were diagnosed within
48 h of hospital admission (community-onset) based on positive MRSA culture and clinical
presentation were equally likely to be infected with high MIC strains compared to those beyond
43
48 h (nosocomial acquisition), 33% (41/129) vs 31% (16/51). Nearly half of our patients were
infected with PVL-positive strains (81/180, 45%). All PVL (+) isolates had SCCmec type IV
cassette. High vancomycin MIC was significantly associated with absence of PVL genes (38%
vs 23%, p=0.008) and SCCmec type II (41% vs 22% SCCmec IV, p=0.019). However, patients
infected with PVL (+) strains were equally likely to meet CDC criteria for CA-MRSA (45%,
37/83) versus HA-MRSA (55%, 42/83). Thus, based on these results molecular markers for CA-
MRSA seem to be good markers for predicting vancomycin MIC that in turn be used to guide
empiric dosing of vancomycin.
While susceptibility to vancomycin was predictable by SCCmec type and PVL status
within our 2005-2007 cohort of patients, we found analyzing vancomycin MICs from 2008-2011
separately, predictability was no longer possible (Figure 3). Initially in 2005-2007 only 31% of
isolates had vancomycin MICs >1µg/ml, however this increased to 52% in our 2008-2011
cohort.
Figure 3. Distribution of Vancomycin MIC by SCCmec Type among 2005-2007 vs 2008-
2011 patients.
IV/V
II
0
20
40
60
0.5
0.75
1
1.5
2
2005-‐2007
IV/V
II
44
Figure 3. Depicts the change in SCCmec IV and PVL status in two different periods of time.
Discussion.
Three clinical studies were conducted from 2005-2011. One was a broad study examining
all MRSA isolates collected from 2005-2007. The remaining two studies were disease specific,
examining MRSA bacteremia and pneumonia within our institution from 2005 - 2011. The
purpose of these studies was to investigate the molecular epidemiology of MRSA within our
institution and identify patient and bacterial characteristics that can potentially be used to guide
treatment decisions.
Few studies have investigated the epidemiology of MRSA infections among hospitalized
patients within the Los Angeles area, although outbreaks by CA-MRSA strains have been
reported (2003a; 2003b; Barrett and Moran, 2004). Other studies have reported in different parts
of the country that CA-MRSA strains have become a significant cause of infections within
hospitalized patients, in some studies these strains were responsible for over 33% of infections
(Davis et al., 2006; Maree et al., 2007; Zervos et al., 2008). We found at our institution located
within Los Angeles County, that CA-MRSA strains accounted for approximately 50% of MRSA
IV/V
0
10
20
30
40
50
60
0.5
0.75
1
1.5
2
>2
2008-‐2011
IV/V
II
45
infections from 2005-2011. During the period of 2005 – 2007 CA-MRSA strains had only
accounted for ~33% of all infections. However, when the time interval was extended from 2005
– 2011, CA-MRSA strains accounted for 50% of invasive infections, which included bacteremia
and pneumonia infections. When excluding patients from 2005 – 2007, it was found that CA-
MRSA strains still accounted for ~48% of MRSA infections from 2008 - 2011. This is consistent
with the above reports that found increasing numbers of infections among hospitalized patients
being attributed to CA-MRSA, suggesting a change in MRSA epidemiology. Based on our
findings and those of other investigators, it appears that CA-MRSA strains are infiltrating the
healthcare setting as we found significant portion (~40%) of strains that would have been defined
as HA-MRSA by the CDC clinical criteria were in fact CA-MRSA strains by genotype.
Consistent with published guidelines and current practice at other institutions ([Anon],
2005; Klevens et al., 2007; Rybak et al., 2009), vancomycin was the empiric and directed
therapy of choice in our studies, where over 85% of patients received it. In practice, patients who
reside at home prior to admission, acquired the infection in the community, or have no prior
healthcare exposure risk may be presumed to be less likely infected with high MIC strains due to
the low likelihood of prior exposure of strains to vancomycin. Notably, nearly one third of the
infected strains in our study conducted from 2005-2007 had high vancomycin MIC. We found
that patients hospitalized for MRSA infections were equally likely to be infected with high
vancomycin MIC strains regardless of place of acquisition and healthcare exposure risk
including those patients coming from the community settings. This observation is consistent with
others who reported an increase in CA-MRSA strains causing infections in hospitalized patients
and HA-MRSA strains infecting patients in the community (Davis et al., 2006; Maree et al.,
2007). The mixing of CA- and HA-MRSA strains across healthcare settings has blurred the
46
epidemiologic distinctions between two MRSA genotypes and obviated the usefulness of these
clinically based parameters in predicting antibiotic resistance phenotypes and thus guiding
empiric treatment decisions.
This decrease in usefulness of clinical characteristics in differentiating S. aureus
susceptibility patterns had previously been demonstrated with methicillin resistance among
MSSA infections (Miller et al., 2007), however we have extended these findings to MRSA and
specifically with the treatment of vancomycin. Furthermore, we demonstrated that while clinical
characteristics were not predictable of vancomycin susceptibility, bacterial epidemiological
markers were predictive. The presence of genes encoding PVL and SCCmec IV and V cassettes,
which have been strongly linked to MRSA strains originating from the community, with ~90%
of strains possessing these genes (Naimi et al., 2003; Zetola et al., 2005). Hence, we investigated
the utility of PVL genes and SCCmec type as genetic markers to differentiate MRSA strains with
reduced susceptibility to vancomycin and found that the odds of a PVL-negative, SCCmec II
MRSA strain having high vancomycin MIC is at least 2-fold higher than a PVL-positive,
SCCmec IV strain (OR 2.4, 95% confidence interval 1.14 to 2.93). Unfortunately, in our later
cohort of patients from 2008-2011 vancomycin MIC was no longer predictable by molecular
markers. This most likely is due to the shift in CA-MRSA epidemiology, where we found ~50%
of infections within the healthcare setting were caused by these strains. Once these CA strains
enter the healthcare system they are likely to undergo the same pressure as HA, resulting in
increase vancomycin MICs. However, it is still possible that in other institutions where CA
strains have not become as large of a burden, that vancomycin MIC may still be predicable by
molecular markers and can still possibly play a role in empiric treatment selection.
Finally, our findings from the analysis of our MRSA bacteremia patient cohort support
47
the IDSA guidelines recommendation to initiate treatment with vancomycin and to assess
continued therapy with vancomycin based on host and microbial status. More importantly, we
provide a timeframe by which a change to alternative therapy based on early clinical response
could potentially affect patient outcomes. We showed that vancomycin continues to be an
effective agent for the empiric treatment of MRSA bacteremia and that clinical response 72h
after initiation of vancomycin therapy is a strong indicator of eventual outcome. Patients that
achieved early response to vancomycin therapy and continued until the end of treatment had the
most favorable outcomes, followed by patients that did not initially respond at 72h but were
switched to a vancomycin alternative. Those who did not respond within 72h after vancomycin
initiation and continued on vancomycin therapy had the worst outcomes as these patient were
more likely to require ICU admission and were 25 times more likely to fail treatment overall.
Our findings are consistent with results from the recent study by Moore et al., which suggest
improved survival in patients not responding to vancomycin initial therapy who were switched to
daptomycin at a median time of 5 days (IQR 3-9 days) (Moore et al., 2012).
In these three clinical studies we have demonstrated shifts in the molecular epidemiology
of MRSA infections, where CA-MRSA strains are now causing healthcare associated disease
and vice versa. CA-MRSA strains were responsible for half of the MRSA invasive infections
investigated. This change effects the clinical judgment used to assess patient empiric therapy
based on clinical criteria, however with the advent of rapid diagnostics, we found it is possible to
use MRSA markers (PVL and SCCmec type) as a guide to empiric vancomycin treatment
dosing. We also have provided a timeframe in which clinical response should be evaluated
during treatment. Day 3 of infection in our bacteremia patients proved to be a significant time
point where therapy should either be continued or re-evaluated for alternative agents. Overall,
48
these three studies provide new approaches that can be utilized to improve the treatment of
patients infected with MRSA.
This chapter was based on the following publications:
1. Yamaki J, Lee M, Shriner K, Wong-Beringer A. Can clinical and molecular epidemiologic
parameters guide empiric treatment with vancomycin for methicillin-resistant Staphylococcus
aureus infections? Diagn Microbiol Infect Dis 2011;70:124-130
2. Yamaki J, Minejima E, Nieberg P, Wong-Beringer A. Molecular epidemiology and in-vitro
production of protein A and alpha-hemolysin amongst MRSA strains causing community-onset
pneumonia. (In preparation)
3. Joo J, Yamaki J, Lou M, Hshieh S, Chu T, Shriner K, Wong-Beringer A. Early response
assessment to guide therapy for methicillin-resistant Staphylococcus aureus bacteremia. Clin
Therapeutics 2013 (Conditionally accepted, Clinical Therapeutics 4/13)
49
Chapter 3.
Contribution of resistance and exoprotein production by MRSA strains
to severity of infection and outcomes
It is well documented that strain-specific virulence contributes to infection with varying severity
in a number of different types of bacterial infections. One well known example is Escherichia
coli serotype O157:H7, which is hyper-virulent compared to other E. coli serotypes, and is
referred to as enterohemorrhagic E. coli. Typically, 10
3
– 10
6
organisms are required to establish
infection with the common E. coli strains, whereas with O157:H7 as few as 10 - 100 can lead to
much more severe infection. The hyper-virulent phenotype exhibited by O157:H7 is due to the
presence of the Shiga toxin (Stx2) that is homolgous to Shiga toxin produced by Shigella
dystenteriae type 1 (Law, 2000; Wang et al., 2002; Werber et al., 2003). This example
demonstrates the increased pathogenesis and virulence of a bacterial strain due to acquisition of
toxin genes not present in typical strains of E.coli (Khan et al., 2011). Clostridium difficile is
another example organism that has been described as varying in virulence due to toxin
production. This example differs from E.coli O157:H7 in that rather than acquiring a toxin
through horizontal gene transfer, C. difficile BNAP/027 strains over-produce toxins A/B that are
already present in most C. difficile strains and the severity of disease and poor patient outcomes
associated with BNAP/027 strains is due to this hyper-production of toxins A/B (Matamouros et
al., 2007). Based on these examples of toxin-mediated disease, we hypothesized that disease
severity and outcomes in humans may also be associated with hyper-virulence of MRSA strains,
which is further supported by the animal models of MRSA infection (Ragle and Bubeck
Wardenburg, 2009; Wang et al., 2007; Wardenburg et al., 2007).
50
Specifically, pneumonia and bacteremia account for 89% of MRSA invasive infections,
of which 12% were community-onset or involving USA300 strains originated in the community
(Klevens et al., 2007). Notably, necrotizing pneumonia caused by community-associated (CA)
MRSA strains carrying the PVL genes has been reported in healthy individuals without apparent
healthcare risks and carries a high mortality rate (Francis et al., 2005; Hunt et al., 1999),
suggesting that CA-MRSA strains may be more virulent than their healthcare-associated (HA-
MRSA) counterparts (Gillet et al., 2002). MRSA virulence in skin soft tissue infections (SSTIs),
pneumonia and bacteremia has been attributed to exoproteins, specifically α-hemolysin (Hla) and
α-type phenol soluble modulins (PSMα
1-4
), which are produced by nearly all S. aureus strains
and in excess by CA-MRSA strains (Li et al., 2009). These exoproteins not only cause direct
damage to target host cells but also exacerbate host inflammatory response, contributing to acute
lung or other nearby tissue injury. Additionally, recent reports have demonstrated the importance
of Protein A (Spa) in pneumonia. All three of these virulence factors are under the control of the
accessory gene regulator (agr), where production of Hla and PSMα
1-4
peak at the time of
maximal agr expression during post-exponential and early stationary phase, while Spa
expression is maximal during exponential growth when agr activity is minimal (Benito et al.,
2000; Novick et al., 1993).
Adding to the challenge of the emerging virulent MRSA clones is the reduced
susceptibility of clinical isolates to vancomycin. Vancomycin has been the gold standard of
treatment since its introduction, and as MRSA infections have increased over the last decade, its
use has also dramatically increased since it is commonly used as directed therapy and empiric
therapy for patients with MRSA risk factors. Thus, there is concern that the increase use over the
years has led to increasing minimum inhibitory concentrations and resistance among MRSA
51
strains (Ho et al., 2010; Wang et al., 2006). Currently, controversy exists over the role of
increased vancomycin MIC (>1 ug/ml) with persistence and failure in bacteremia, where some
studies have found associations while other have not (Musta et al., 2009; Sakoulas et al., 2004).
There is even less knowledge about the contribution of elevated vancomycin MIC in pneumonia,
however a recent study by Haque et al found that the risk of mortality in patients with MRSA
HAP, VAP, and HCAP increased as a function of the vancomycin MIC (Haque et al., 2010).
With the growing concern of vancomycin failures in the treatment of bacteremia and
pneumonia (Haque et al., 2010; Sakoulas et al., 2004) and the increasing incidence of CA-MRSA
strains within the healthcare settings being reported in other parts of U.S. (Davis et al., 2006;
Seybold et al., 2006), we hypothesized that in-vitro increased exotoxin production, increased
cytotoxicity against host target cells, and decreased vancomycin susceptibility of these strains
would be associated with poor patient outcomes and disease severity.
Materials and Methods.
Bacterial Strains. MRSA study isolates included 155 strains that caused pneumonia (134
responsible specifically for community-onset pneumonia), 10 bloodstream isolates, and 50
unique PVL (+) MRSA (n =35) and MSSA (n=15) isolates that caused complicated skin soft
tissue infections in adults patients enrolled in Phase II clinical trial of TR-701. Presence of luk-
FS and SCCmec genes was determined by PCR. Methicillin resistance was confirmed using
VITEK and oxacillin agar screen test (MHA containing 2% added NaCl with 6 mg/L OXA). All
control strains with the exception of LAC were obtained through the Network of Antimicrobial
Resistance in Staphylococcus aureus program supported by the NIIAID/NIH (NARSA.net);
NRS 100 (USA500), NRS 22 (USA600), NRS 144, NRS 155, NRS 149, NRS 384 (USA 300),
52
NRS 123 (MW2), NRS 483 (USA 1000), and NRS 484 (USA 1100). Primer sequences and
conditions were used according to previously published studies (Boye et al, Bonnstetter et al,).
Collection of Supernatants. All isolates were grown overnight at 37°C in a 5ml pre-culture of
Tryptic Soy Broth (TSB) with shaking at 250rpm. The pre-culture was subsequently used to
inoculate a subsequent 5ml overnight culture grown at 37°C with shaking at 250 rpm for
measurement of baseline production of PSMα
1-4
peptides, Hla, and Spa. Supernatants were then
spun down and passed through a 0.32 micron filter to remove bacterial cells and stored at -80°C
until needed.
LC/MS/MS. With collaborators at City of Hope Hospital, PSM peptides were quantitated by the
following methods. Concentration of formylated PSMs alpha 1, 2, 3, and 4 in TSB culture media
were determined using an UPLC-tandem mass spectrometric assay. Solvents and assay reagents
were purchased from Fisher Scientific (Madison, WI). PSMs alpha 1, 2, 3, 4, alpha 1 D5
(internal standard), and PSM alpha 3 D5 (internal standard) were synthesized by The City of
Hope Medical Center. Instrumentation consisted of a Waters Acquity UPLC system in line with
a Waters Quattro Premier XE Triple Quadrupole Mass Spectrometer (Waters, Milford, MA). The
detector settings were as follows: capillary voltage, 4.8 kV; cone voltage for PSM alpha 1, 2, 3,
4, and PSM A3 D5 were 75 V, 78V, 80V, 75V, and 85V, respectively; collision cell voltage for
PSM alpha 1, 2, 3, 4, and PSM A3 D5 were 90 eV, 88 eV, 105 eV,88 eV, and 95 eV,
respectively; source temperature, 125°C; desolvation temperature, 450°C; cone gas flow, 80
liter/h; and desolvation gas flow, 700 liter/h. The mass transitions monitored for PSMs alpha 1,
2, 3, 4, alpha 1 D5 and PSM alpha 3 D5 were 1144.5→86.3, 1153.56→86.3, 1318.22→120.2
53
1100.46→86.3, 1147.85> 86.3 and 1320.15→124.94, respectively. Chromatography consisted of
gradient separation across a Jupiter 4µProteo 90A 150 ×2.0 mm analytical column (Phenomenex,
Torrance, CA) using mobile phase A: 0.1% TFA in water, and B: 0.1% TFA in acetonitrile. The
column temperature was 30 °C and the following gradient program was used: 60% B (0-2 min),
85% B (2.1 min), 94% B (6.5 min), 60% B (6.6 min), 60% (9 min). The flow rate was 0.3
ml/min. The total run time was 9 minutes, and retention times for PSMs alpha 1, 2, 3, 4, and
PSM alpha 3 D5 were 4.07, 3.97, 3.48, 5.07, and 3.48 minutes, respectively. The standard curve
of quantitation was from 0.1 µg/ml to 10 µg/ml.
Western Blotting. Hla and Spa production was assayed using western blot analysis by
precipitating 1ml of supernatant with TCA and concentrating 10-fold. Equal amount of sample
from each condition were then heated at 95̊C for 6 minutes in Laemmli buffer and then run on a
12.5% SDS-PAGE gel (BioRad, Hercules, CA). Transfer to nitrocellulose was performed at
400mA for 70min. Blots were blocked with 5% BSA for 60 min and then incubated with HRP-
labeled anti-Hla (Abcam, Cambridge, MA) overnight at 4°C at a 1/1000 dilution. For western
blot detection of Spa, anti-Protein A antibodies (Abcam, Cambridge, MA) were added at a
dilution of 1/5,000 for 2 hours, washed with PBS-tween20 0.05%, and incubated with secondary
HRP antibodies at a 1/20,000 dilution. Hla (1.5 µg/ml) was used as an internal standard to take
into account any differences in exposure times among blots and used for normalization when
quantifying pixel density by Un-Scan-It densitometry software (Silk Technologies, Orem, UT).
Neutrophil and A549 Cytotoxicity. Cytotoxicity of neutrophils was determined by an LDH
release assay (Promega, Madison, WI). Neutrophils (PMNs) were isolated from healthy
54
volunteers under IRB-approved protocol. One-Step polymorph per manufacturer protocol was
used for neutrophil isolation followed by plating 10
6
neutrophils per well in a 96-well plate. Two
control strains (LAC and USA600) and a total of 20 MRSA clinical strains were selected for
cytotoxicity experiments based on SCCmec type, PVL status, severity of pneumonia presentation
and patient outcomes. Strains were grown overnight in TSB at 250 rpm at 37C and then diluted
to give a final M.O.I. of 50. PMNs were then incubated for 4 hours with each strain. Cytotoxicity
against A549 lung epithelial cells was performed as described previously (Wardenburg et al.,
2007). Briefly, S. aureus strains were grown overnight in TSB at 37°C at 250rpm, then re-
inoculated and grown to mid-exponential log phase (OD
600
= 0.4). Cells were then pelleted and
washed once with PBS before re-suspending the pellet in 10 ml F12K media. 100µl of bacteria in
F12K were then added to a 96-well plate that was seeded overnight (18 h) with 15,000 A549
cells per well.
Statistical Analysis. Continuous variables were compared by student t-test or Wilcoxan Rank
Sum test. Variation of PSMα peptide concentrations compared by ANOVA with Dunns
correction by GraphPad Prism 5.0 software.
Results.
Baseline production of PSMα’s in various types of infection.
PSMα production in SSTI and contribution to wound severity. 50 clinical isolates that caused
cSSSIs were tested. Nine laboratory MRSA control strains were also analyzed and compared for
PSMα production. The measured PSMα concentrations among clinical isolates ranged from 0.22
55
to 98.24 µg/ml, with PSMα
1
and α
4
produced in greater concentrations than PSM α
2
and α
3
(Figure 1).
Figure 1. PSMα peptide production among S. aureus clinical isolates causing cSSSI.
Production of PSMα peptides from clinical S. aureus (MSSA and MRSA) isolates varied among strains
and by PSMα peptide. Mean production of PSMα
1-4
was 26.32, 13.9, 13.2, and 32.3 µg/ml, respectively.
When compared to the LAC control strain, some clinical isolates produced up to at least 2.5
times higher amounts of all four PSMα peptides. As expected, the agr null strain (RN9120)
produced no measurable PSMα’s and the partial agr defective mutant (RN4220) produced
minimally measurable amounts of PSMα
4
. Control strains that were community-associated (CA-
MRSA) USA300, USA1000, USA1100, and WIS produced higher amounts of PSMα’s
compared to those that are of the healthcare-associated (HA-MRSA) background, COL, 85/2082,
and USA600. Clinical isolates also followed the pattern of the control strains, as CA isolates
produced higher amounts of all PSMαs compared to HA strains (Figure 2a). This is consistent
with previously published results (Li et al., 2010; Yamaki et al., 2011b). Similar concentrations
of PSMα peptides were produced among clinical isolates regardless of methicillin resistance,
type of skin and skin structure infection (cellulitis with or without abscess) or size of the abscess
(> 5 or ≤ 5cm) caused by these strains (Yamaki et al., 2011b).
PSM α1
PSM α2
PSM α3
PSM α4
0
20
40
60
80
100
PSM α4
PSM α1
PSM α2
PSM α3
PSMα µg/ml
56
Figure 2. Variation in PSMα peptide production by PVL status and invasiveness.
a) b)
1
α
PSM
2
α
PSM
3
α
PSM
4
α
PSM
0
10
20
30
40
PVL (-)
PVL (+)
**
*** ***
***
PSM
µ
g/ml
1
α
PSM
2
α
PSM
3
α
PSM
4
α
PSM
0
10
20
30
40
Non-invasive
Invasive
***
*
*
PSM
µ
g/ml
PSMα production categorized as PVL (+) or (-), all PVL (+) strains were SCCmec IV and all PVL (-) were
SCCmec II, characteristic of CA vs HA strains. (a) Production of PSMα peptides by clinical strains
grouped by PVL status. PVL (+) strains produced significantly higher amounts PSM peptides compared to
PVL (-) strains. (b) PSMα production levels also varied by invasiveness of disease. Where higher amounts
were produced by strains causing non-invasive disease. (* p < 0.0228, ** p = 0.0016, *** p < 0.00081)
PSMα contribution in isolates causing invasive disease. Thirty-one invasive clinical isolates (10
bacteremia, 21 pneumonia) were tested for PSMα production to determine if any differences in
PSMα production at baseline may contribute to invasiveness. We hypothesized that invasive
strains would produce higher amounts of PSMα peptides, giving these strains an advantage in
evading the immune system, aiding in nutrient acquisition, and disseminating, thus leading to
invasive diseases. Significant differences were observed between PSMα production of invasive
vs non-invasive strains, however we found that non-invasive strains in fact produced
significantly higher amounts of all four PSMα peptides compared to invasive strains (Figure 2b).
Additionally, no significant differences were observed between the two categories of invasive
strains, pneumonia vs bacteremia in PSMα peptides produced. With the invasive group of
infections no correlation was found between patient outcomes and PSMα production, likely due
to the small sample size of each disease state.
57
Baseline production of α-hemolysin and Protein A in community-onset pneumonia.
As MRSA is an increasingly important cause of pneumonia, we sought to examine the
contribution of two other exotoxins, α-hemolysin (Hla) and protein A (Spa) in pneumonia. These
toxins have been demonstrated in animal models to play a more critical role in pneumonia than
PSMα peptides (Bubeck Wardenburg and Schneewind, 2008; Diep et al., 2013; Gomez et al.,
2004; Soong et al., 2011a). To investigate the contribution of these toxins in pneumonia, we
included 134 patients and quantitated the amount of Hla and Spa for each strain and selected a
subset representative of patient outcomes to test their cytotoxicity against host target cells.
Immunoblot assays were performed on all 134 isolates for Hla and Spa production. Pixel
density, after normalizing to an internal standard of recombinant Hla and controlling for CFUs of
overnight cultures was used for analysis. Most clinical isolates produced Hla and Spa; 30 isolates
(22%) had no detectable Hla levels, with 71% of these strains being PVL (-) and only 9 isolates
(7%) did not produce any detectable Spa, seven of which were PVL (-). Confirmation of Hla
absence was confirmed by plating on blood agar plates, where in all cases hemolysis was
minimally discernable suggesting Hla was produced below the limit of detection by the
immunoblot or other exoproteins were responsible for the slight RBC lysis.
Protein A was more abundantly produced among strains compared to α-hemolysin in all
strains, except in strains where Spa was not produced at all. As mentioned earlier, a number of
studies using laboratory strains or small number of clinical strains have demonstrated increases
in exotoxin production in CA compared to HA-MRSA strains. Hence, we analyzed production of
Hla and Spa based on CA vs HA and the presence of PVL-encoding genes and SCCmec types. In
both comparisons, production of Hla was significantly higher in PVL (+) and SCCmec IV strains
58
compared to strains that were PVL (-) or non-SCCmec IV/V strains (p <0.0001). Spa production
also differed between these groups, however not as significantly as Hla (p = 0.0109) (Figure 3).
Figure 3. In-vitro α-hemolysin and protein A production based on PVL status.
a)
b)
PVL +
PVL -
0
1
2
3
4
P value 0.0109
*
Pixel density normalized by CFU
Strains were grouped by PVL status as a marker for CA vs HA, and production of Hla (a) and Spa was
compared (b). Both Hla and Spa were significantly produced in higher amount by PVL (+) strains
compared to PVL (-), the same is true when compared by SCCmec IV vs II.
In-vitro Hla and Spa production and correlation with patient outcomes. In examining the
relationship between disease severity and Hla or Spa production, we compared pneumonia
severity index scores, if mechanical ventilation was required, whether pneumonia was mono or
multi-lobar, or if infection resulted in sepsis. When analyzing severity of presentation based on
mean production of Hla and Spa, we found no association between in-vitro level of production
and severity (Table 1). Notably, strains isolated from patients who suffered from sepsis produced
less Hla compared to strains from patients without sepsis (0.63 µg/ml (0, 2.8) vs 2.2 µg/ml
(0.6,3.5), p = 0.024).
PVL +
PVL -
0
1
2
3
4
***
P value < 0.0001
Hla µg/ml normalized by CFU
59
Table 1. Disease severity and outcomes based on Hla and Spa production.
Outcome variable Hla ug/ml p-value Spa density p-value
Sepsis Yes = 0.63 (0, 2.8)
No = 2.2 (0.6, 3.5)
0.024 Yes = 2.79 (1.3, 3.89)
No = 2.62 (1.14, 3.88)
0.1541
Re-admit for MRSA
pneumonia
Yes = 2.6 (1.22, 4.31)
No = 2.58 (1.17, 3.86)
0.94 Yes = 1.93 (0.61, 3.32)
No = 2.84 (1.27, 4.02)
0.1919
PSI Score Yes = 1.81 (0.39, 3.65)
No = 2.32 (0.33, 3.93)
0.30 Yes = 1.9 (1.22, 2.93)
No = 2.64 (1.18, 3.89)
0.2651
Mechanical
ventilation
Yes = 1.80 (0.39, 3.47)
No = 2.01 (0, 3.39)
0.7038 Yes = 1.99 (1.15, 3.65)
No = 2.85 (1.22, 4.04)
0.27
Multi-lobar
pneumonia
Yes = 1.48 (0.29, 3.33)
No = 1.93 (0.28, 3.46)
0.6649 Yes = 2.23 (0.98, 3.75) No
= 3.07 (1.4, 4.44)
0.0248
Respiratory failure Yes = 2.63 (0.46, 3.95)
No = 1.77 (0.04, 3.38)
0.7388 Yes = 1.99 (1.08, 3.73)
No = 2.85 (1.13, 4.01)
0.5401
Days until stable
<=7d vs 7d
Yes = 1.33 (0, 3.34)
No = 2.5 (0.57, 3.51)
0.2916 Yes = 3.1 (1.45, 3.52)
No = 2.84 (1.54, 4.43)
0.5310
ICU admit Yes = 1.78 (0, 3.57)
No = 2.32 (0.46, 3.38)
0.767 Yes = 2.58 (1.32, 3.82)
No = 2.85 (0.97, 4.04)
0.883
30-day mortality Yes = 2.27 (0.30, 3.91)
No = 1.65 (0.29, 3.31)
0.32 Yes = 2.59 (1.78, 4.16)
No = 2.6 (1.16, 3.8)
0.469
NOTE: Yes – Patients who had the outcome variable, No – patients who did not have the
outcome variable. Values are reported as median (IQR)
Similarly with respect to patient outcomes, length of hospital stay or length of ICU stay, early
death (within 1 week) and 30-day mortality, no significant association between production level
of Hla and Spa and outcomes was observed. Furthermore, there was no significant difference
between the production levels of Hla in strains isolated from patients who were re-hospitalized
within 30 days due to MRSA pneumonia.
Cytotoxicity of selected clinical strains. Cytotoxicity of 19 clinical isolates was tested separately
in cell cultures against two host cell types: A549 lung epithelial cells and human neutrophils.
Isolates were selected based on SCCmec type (IV vs II), presence of PVL genes, Hla and Spa
production (high vs none), and patient outcomes including: respiratory failure, PSI, bilateral pna,
and if the patient expired due to MRSA. In-vitro cytotoxicity differed by SCCmec type and
depending on host cell type. Interestingly, human PMNs were more susceptible to strains that
60
were CA-MRSA compared to HA-MRSA strains, where the mean percent cytotoxicity of CA
strains was 40% vs 16.5% for HA-MRSA strains (p = 0.02) (Figure 4).
Figure 4. Differential cytotoxicity of clinical MRSA strains against A549 and human PMNs.
a)
b)
II
IV
0
20
40
60
*
P value 0.0209
PMNs
SCCmec type
% cytotoxicity
II
IV
0
20
40
60 P value 0.0279
*
A549
SCCmec type
% cytotoxicity
Clinical MRSA strains (n=19) displayed differential cytotoxicity against human PMNs (a) and A549 lung
epithelial cells (b) depending on the strain SCCmec type. All SCCmec IV strains were also PVL (+)
indicating CA-MRSA strains. SCCmec II strains were significantly more cytotoxic against A549 cells (b)
while SCCmec IV were more cytotoxic against PMNs (a).
However, HA-MRSA strains produced significantly more A549 lysis compared to CA strains,
the mean percent cytotoxicity for HA strains was 31% vs 18% for CA strains (p = 0.0279).
Cytotoxicity of A549 and human PMNs did not correlate with in-vitro Hla or Spa production, as
even strains that did not produce detectable levels of Hla and Spa were able to cause significant
cell lysis. As with patient outcomes, in-vitro Hla and Spa production was not correlated with
cytotoxicity.
Contribution of vancomycin susceptibility to patient outcomes.
So far, we have focused our investigations on the molecular epidemiology of MRSA strains, the
associated toxins produced by these strains, and how these factors influence patient outcomes.
Another important aspect to consider in treatment outcomes of patients with MRSA infections is
the organism’s susceptibility to antibiotics used for treatment. As mentioned, vancomycin is the
61
current gold standard of treatment for MRSA infections, however some studies suggest that there
MICs to vancomycin may be increasing steadily as MRSA infections have become more
common, this has been coined the “vancomycin creep” (Ho et al., 2010; Moise et al., 2009).
There is no consensus as to whether or not “vancomycin creep” is actually occurring or if it is a
phenomenon associated with different methods in measuring susceptibility. However, it is
thought that strains with decreases susceptibility to vancomycin may lead to clinical failure,
particularly if strains express the hVISA phenotype (Moise et al., 2009; Musta et al., 2009; van
Hal and Paterson, 2011).
In chapter 2 we demonstrated in the time period from 2005-2007, susceptibility to
vancomycin (MIC ≤ 1µg/ml) could be predicted by molecular markers associated with CA-
MRSA (SCCmec IV/V and PVL genes). Likewise, this was for the most part true for pneumonia
isolates collected at our institution from 2005-2011. However, during this same time period
vancomycin susceptibility was not predictable in isolates collected from patients with
bacteremia, as both CA and HA-MRSA as determined by molecular markers were just as likely
to have vancomycin MICs >1µg/ml. Hence, with the inability to predict vancomycin MIC by use
of molecular markers in bacteremia and the small number of reports describing treatment failure
with vancomycin against MRSA strains with high vancomycin MIC, we analyzed our data to
determine if in our cohort of patients infected with high vancomycin MIC (>1µg/ml) strains and
the hVISA phenotype had poor outcomes.
Relationship between vancomycin susceptibility and hVISA phenotype and outcomes in
bacteremia. In our cohort of patients with bacteremia the distribution of high vancomycin MIC
values were equal between CA and HA-MRSA strains. 70% of all isolates had vancomycin MIC
62
>1µg/ml and strains exhibiting the hVISA phenotype was observed in ~11% of cohort patients.
Surprisingly, the hVISA phenotype was also equally distributed among CA vs HA-MRSA
strains at 10% and 12%, respectively. Outcomes based on vancomycin MICs did not differ
between the high and low MIC groups. Length of stay was similar between low and high MIC
groups (16d vs 14.5d). Treatment failures among this cohort of bacteremia patients were not
significantly predicted by vancomycin MIC >1µg/ml (p = 0.21), this also remained true after
multi-variate analysis (p = 0.0615). The high vancomycin MIC group had twice the number of
deaths (14% vs 7%) however this difference was not statistically significant (p=0.25).
Vancomycin susceptibility and hVISA phenotype associated with outcomes in pneumonia. The
majority (72%) of all pneumonia isolates collected from 2005-2011 had vancomycin MIC values
of >1µg/ml. Vancomycin MICs were not associated with any clinical outcomes including:
mortality, clinical failure, length of stay, time to stability, infection related complications,
readmission, or ICU admissions. Although, patients infected with low vancomycin MIC strains
were less likely to have recent exposure to antibiotics compared to patients with high MIC
strains (p = 0.05) and less likely to have multiple co-morbid conditions (p = 0.03). The hVISA
phenotype was also found in 11% in the study isolates. As with the bacteremia study no
association between high MIC or hVISA and poor outcomes was observed in our study group.
Alternatives to vancomycin therapy for MRSA. In our patient population we did not find any
statistically significant association between high vancomycin MIC and poor outcomes. We did
however find a trend in vancomycin high MIC and failure (p = 0.0615) in our bacteremia
population and considering other investigators have reported an increase in failures associated
63
elevated vancomycin MICs (Moise et al., 2009; Sakoulas et al., 2004) it is of importance to
monitor susceptibility of alternative agents.
Thus, in anticipation for the need to consider alternative parenteral agents (Hidayat et al.,
2006; Sakoulas et al., 2004; Soriano et al., 2008), we determined strain susceptibility to
daptomycin, tigecycline, and linezolid using eTest methodology. 180 clinical isolates tested were
susceptible to all three antibiotics per CLSI interpretive criteria. MIC
50
, MIC
90
, and geometric
mean MIC values for alternative agents were similar when compared by vancomycin MIC (high
vs low), PVL status, and invasiveness of infection caused by these strains (Table 2). Using
Spearman’s correlation to examine the relationship between vancomycin MIC and those of the
other three antibiotics, only a weak but statistically significant correlation was observed between
vancomycin and daptomycin (ρ = 0.225, p = 0.0024) and vancomycin and linezolid (ρ = 0.214, p
= 0.004) (Yamaki et al., 2011a).
Table 2. Comparative susceptibility of MRSA strains to vancomycin, daptomycin, linezolid,
and tigecycline.
VAN
MIC
50
VAN
MIC
90
VAN
Mean
MIC
DPT
MIC
50
DPT
MIC
90
DPT
Mean
MIC
LZ
MIC
50
LZ
MIC
90
LZ
Mean
MIC
TYG
MIC
50
TYG
MIC
90
TYG
Mean
MIC
PVL (+) 1 1.5 1.06 0.25 0.38 0.30 0.75 1 0.82 0.074 0.094 0.06
PVL (-) 1 1.5 1.15 0.25 0.38 0.27 1 1 0.92 0.064 0.094 0.076
Non-inv 1 1.5 1.10 0.25 0.38 0.30 0.75 1 0.83 0.047 0.094 0.065
Invasive 1 1.5 1.12 0.25 0.38 0.27 1 1 0.90 0.064 0.094 0.07
VAN ≤1 - - 0.93 0.25 0.38 0.27 0.75 1 0.85 0.064 0.094 0.068
VAN >1 - - 1.51 0.25 0.38 0.31 1 1 0.93 0.064 0.094 0.069
NOTE: VAN = vancomycin, DPT = daptomycin, LZ = linezolid, TYG = tigecycline, MIC50 = MIC value
for 50% of the isolates, MIC90 = MIC value for 90% of the isolates, Mean MIC = Geometric mean value
of all isolates tested.
The newest classes of antibiotics to have activity against MRSA are the fifth-generation
cephalosporins. Currently, only ceftaroline is FDA approved for the treatment of MRSA
infections. It is currently FDA approved for only SSTI in MRSA infections, however clinicians
may still prescribe ceftaroline as off label use in bacteremia and pneumonia. With this in mind
64
we sought to test the MICs of our pneumonia isolates against ceftaroline to gain insight into its
possible role in the treatment of pneumonia as an alternative agent to vancomycin. We found all
of our pneumonia isolates were susceptible to ceftaroline. The MIC
50
and MIC
90
were both 0.5
µg/ml, these results are similar to those reported by Jones et al where MIC
50
and MIC
90
were
0.25-0.5 µg/ml and 1ug/ml, respectively (Jones et al., 2011). Suggesting that despite 72% of
isolates having vancomycin high MICs, strains remained highly susceptible to ceftaroline. No
correlation was found between vancomycin MIC and ceftaroline MIC by Spearman evaluation (p
= 0.7233).
Discussion.
Literature suggests that both exotoxin production and resistance impact patient outcomes. CA-
MRSA strains are thought to be more virulent than their HA-MRSA counter parts due to
increased toxin production (Hla and PSMα’s) while treatment failures in bacteremia are thought
to be related to vancomycin resistance. While multiple studies have supported the relationship
between vancomycin resistance and treatment failure only one small study has examined the
contribution of exotoxin production on disease presentation in patients infected with MRSA.
Hamilton et al conducted a study with 29 CA-MRSA strains looking at in-vitro production of
PVL of isolates causing pneumonia, bacteremia, and SSTI. No correlation between disease type
or severity and PVL was found (Hamilton et al., 2007). In this study we determined the
contribution of exotoxins (Spa, Hla, PSMs) and resistance may play in infections by using a
large collection of clinical isolates.
PSMα peptide production was measured among 81 strains causing non-invasive and
invasive infections. Significant differences in the level of production of all four PSMα peptides
65
were observed between CA strains compared to HA (p < 0.026), this is consistent with other
reports (Li et al., 2010). The amount of PSMα production did not correlate with wound or
abscess sizes in cSSTIs. Also patient outcomes and severity of infection did not significantly
differ in any disease state (SSTI, pneumonia, bacteremia) based on in-vitro PSMα production.
Interestingly, we did find higher amounts of PSMα peptides being produced by strains causing
non-invasive infections compared to invasive, however this is likely due to the higher number of
CA-MRSA strains in the non-invasive group.
Our study tested the largest collection of clinical CA and HA-MRSA isolates for in-vitro
Hla and Spa production. Similar to results obtained with PSMα peptide production with
outcomes and disease severity, we did not find any association between Spa and Hla production
and severity of disease in our pneumonia patients. We did report significant differences in Hla
in-vitro production among CA vs HA clinical strains, where CA-MRSA strains produced
significantly higher amounts (p < 0.0001). Cytotoxicity assays of selected strains also did not
correlate with outcomes and disease severity, however we did find that the tested clinical isolates
differed in cytotoxicity against different host cell types. Human neutrophils were highly
susceptible to lysis by CA compared to HA-MRSA strains, while A549 lung epithelial cells were
more susceptible to HA strains compared to CA (p < 0.05). To our knowledge, we are the first to
report differences in host cell susceptibility to clinical strains from different molecular types.
We speculate that other contributing factors most likely play roles in disease progression
of MRSA pneumonia and SSTI. Strictly looking at the bacterial side, it is expected that other
toxins produced in concert with the tested toxins likely play roles in disease progression and
severity. Additionally, it is possible that in-vivo toxin production differs from in-vitro, as
production of exotoxins by the same strains can differ by the type of culture media used (Graves
66
et al., 2010) and host environmental cues may lead to toxin production. Host immune response to
the infecting strain may also contribute to disease outcomes and severity. Furthermore, previous
exposure of patients to MRSA and the toxins produced may have led to antibody production,
which would in turn neutralize their effects. Finally, drug treatment is known to affect the
production of a number of toxins, particularly β-lactams which have been demonstrated in-vitro
to induce toxin production including PVL and Hla, thus improper empiric treatment may have
also played a role (Dumitrescu et al., 2007; Stevens et al., 2007a).
As resistance has been demonstrated to adversely affect patient outcomes in a number of
bacteremia studies and in a recent study in MRSA HAP (Haque et al., 2010; Sakoulas et al.,
2004) we examined resistance patterns within our cohort of patients. We found in our cohort of
patients, strains with elevated vancomycin MICs (>1µg/ml) were responsible for ~72% of all
infections. Additionally, in both the pneumonia and bacteremia studies we found 11% of strains
exhibited the hVISA phenotype and this phenotype was equally observed in CA and HA strains.
This observation is important, as CA-MRSA strains have been typically described as more
susceptible to non-β-lactam antibiotics. Contrary to earlier described studies showing poorer
patient outcomes with higher levels of resistance to vancomycin, we did not find any significant
association with vancomycin MICs >1µg/ml or the hVISA phenotype. However, a trend was
observed with vancomycin MICs >1µg/ml and treatment failure (p = 0.06), and a numerically
higher number of deaths in this group in our bacteremia cohort.
To anticipate the eventual decrease in effectiveness of vancomycin, as we have reported a
high percentage of strains with elevated vancomycin MICs and hVISA phenotype, we
investigated susceptibility patterns of alternative agents to vancomycin and any relationship that
may exist between vancomycin resistance and susceptibility to these alternatives. All isolates
67
tested were susceptible to daptomycin, tigecycline, linezolid, and ceftaroline. Only a slight
correlation was found between increasing vancomycin MICs and daptomycin and linezolid MICs
(p = 0.004). While this correlation was statistically significant, it at this point is not clinically
significant as all strains were below the MIC breakpoints for the alternative agents.
In conclusion, resistance was not found to play as large of a role in outcomes as expected
based on previous studies, however we did find increased likelihood of treatment failure and
increased numbers of deaths with high vancomycin MIC strains. We did not find any correlation
between patient outcomes or disease severity with in-vitro exotoxin production in MRSA strains.
While exotoxin production has been shown to play major roles in the pathogenesis of E. coli and
C. difficile, it is likely a number of other factors may obscure the true roles of MRSA exotoxins
during in-vivo human infections. Strong evidence for the roles of PSMα peptides and Hla in
murine and rabbit models of infection still exists, suggesting severe MRSA infections are toxin
mediated. This is corroborated by studies of vaccines targeting toxins such as Hla and PVL and
decreasing disease severity (Brown et al., 2009; Bubeck Wardenburg and Schneewind, 2008).
This chapter was based on the following manuscripts:
1. Yamaki J, Synold T, Wong-Beringer A. Anti-virulence potential of TR-700 and clindamycin
on clinical isolates of Staphylococcus aureus producing phenol-soluble modulins. Antimicrob
Agents Chemother. 2011 Sep;55(9):4432-5
2. Yamaki J, Minejima E, Nieberg P, Wong-Beringer A. Molecular epidemiology and in-vitro
production of protein A and alpha-hemolysin amongst MRSA strains causing community-onset
pneumonia. (In preparation)
3. Joo J, Yamaki J, Lou M, Hshieh S, Chu T, Shriner K, Wong-Beringer A. Early response
assessment to guide therapy for methicillin-resistant Staphylococcus aureus bacteremia. Clin
Therapeutics 2013 (Conditionally accepted, Clinical Therapeutics 4/13)
68
Chapter 4.
Effects of protein synthesis inhibitors on exotoxin modulation in MRSA, in-vitro and in a
cellular model of infection
MRSA virulence in skin soft tissue, pneumonia, and bacteremia has been attributed to
exoproteins, specifically α-hemolysin (Hla) and α-type phenol soluble modulins (PSMα
1-4
)
which are produced by nearly all S. aureus strains and in excess by CA-MRSA strains (Li et al.,
2009). These exoproteins not only cause direct damage to target host cells but also exacerbate
host inflammatory response, contributing to acute lung injury. Additionally, recent reports have
demonstrated the importance of Protein A (Spa) in pneumonia. Spa binds to the Fc portion of
antibodies allowing for evasion of host immune system and aids in spreading through epithelial
barriers within the host by altering tight junctions of epithelial cells (Martin et al., 2009; Soong et
al., 2011b). All three of these virulence factors are under the control of the accessory gene
regulator (agr), where production of Hla and PSMα
1-4
peak at the time of maximal agr
expression during post-exponential and early stationary phase, while Spa expression is maximal
during exponential growth when agr activity is minimal (Benito et al., 2000; Novick et al.,
1993).
Adding to the challenge of the emerging virulent MRSA clones is the reduced
susceptibility of clinical isolates to the standard treatment option, vancomycin. In light of the
impressive arsenal of virulence factors contributing to the success of MRSA as a pathogen and
increasing MICs to vancomycin, it is of keen interest to determine if alternative agents belonging
to the antibiotic class of protein synthesis inhibitors exert added anti-virulence benefit of
exoprotein inhibition. Clindamycin, linezolid, and tigecycline are commonly prescribed anti-
MRSA agents belonging to this class of antibiotics. Results from previous studies evaluating a
69
limited number of isolates have shown that linezolid and clindamycin markedly decrease
production of Panton-Valentine leukocidin (PVL), Hla, and PSMα
1-4
(Dumitrescu et al., 2008;
Yamaki et al., 2011b), though the effect of tigecycline on exoprotein production has not been
studied. Based on the above observations, some clinicians make the general assumption that all
protein synthesis inhibitors have the added benefit of inhibiting exoprotein production in addition
to bacterial growth.
In the present study, we investigated the anti-virulence potential of protein synthesis
inhibitors commonly prescribed for the treatment of MRSA infections and a second-generation
Oxazolidinone currently in phase III trials. Our goal was to determine whether anti-virulence
effects can be generalized across different clinical isolates, different agents that inhibit protein
synthesis, and different key exoproteins produced by MRSA clinical strains. Specifically, we
tested the effect of clindamycin, linezolid, tigecycline, and tedizolid (TR-700) at sub-inhibitory
concentrations (½, ¼, and ⅛ MIC) on formylated PSMα
1-4
, Hla, and Spa production among
MRSA isolates representative of SCCmec II and SCCmec IV types causing pneumonia and
bacteremia. mRNA expression of psm, hla, spa, as well as mRNA of the global regulators
RNAIII and agrA were also quantitated under these conditions as the agr system is known to
positively control expression of PSMα peptides. As sub-inhibitory protein synthesis inhibitors
have been previously shown to decrease Hla and global regulators expression (RNAIII), we
hypothesized that sub-inhibitory protein synthesis inhibitors would also decrease PSMα peptides,
Hla and Spa production, decrease global regulator expression, and decrease cytotoxicity under
these conditions.
70
Methods.
Bacterial Isolates. A total of 40 MRSA isolates were used in this study (13 invasive clinical
isolates, 24 SSTI clinical isolates, and 3 control strains). Control strains (USA400 and USA600)
and LAC psm and hla deletion mutants were obtained from the Network on Antimicrobial
Resistance in S. aureus (NARSA), and LAC (USA300) was obtained from the Los Angeles
County Public Health Department. All clinical MRSA isolates tested in the study were obtained
from patients hospitalized at Huntington Hospital confirmed to have invasive infections
(bacteremia or pneumonia and one necrotizing fasciitis). Informed consent was waived by the
institutional IRB as all data have been de-identified and no intervention was made.
All isolates were grown overnight at 37°C in a pre-culture of Tryptic Soy Broth (TSB)
with shaking at 250rpm. The pre-culture was subsequently used to inoculate an overnight culture
grown at 37°C with shaking at 250 rpm for measurement of baseline production of PSMα
1-4
peptides, Hla, and Spa.
Plasmid Construction. The plasmid pAMIlux was generously donated form Dr. Julian Davies
laboratory (Diagram 1). The plasmid contains a modified lux operon that is specifically designed
for use in gram (+) bacteria. When obtained, the lux operon on the plasmid was promoter less,
thus the psmα promoter was cloned and inserted. Primers for amplifying the psmα promoter were
created based off of sequences for USA300 at NIH/NCBI and were designed to have BamHI
restriction sites. PCR was carried out to amplify the product. Primer sequences can be found in
(Table 1). The psmα promoter PCR product was purified by spin column (Qiagen, Valencia,
CA). Then both the PCR product and pAMIlux plasmid were treated for 2h at 37°C with BamHI
restriction enzyme (NEB, Ipswich, MA) per manufacturer protocol. Digests were then treated
71
with calf intestinal phosphatase for 1h at 37°C (NEB, Ipswich, MA) and spun through a mini
prep column (Qiagen, Valencia, CA) to remove excess enzyme before ligation. T4 DNA ligase
(NEB, Ipswich, MA) was used to ligate the psmα promoter product and pAMIlux plasmid at
room temperature for 30 minutes per manufacturer protocol.
Diagram 1. Improved lux reporter for use in S. aureus
Plasmid from Dr. Julian Davies laboratory used for creation of psmα promoter reporter.
Electroporation. Electroporation was conducted following a previously optimized protocol for
Staphylococcus carnosus. (Lofblom et al., 2007). Briefly, electrocompetent cells were generated
by diluting an overnight culture of S. aureus into 500mls of B2 media to an OD
600nm
of 0.2 and
allowed to grow until an OD of 0.6. Cells were chilled on ice for 15 minutes and then centrifuged
at 3000g for 10 minutes at 4°C. Cells were washed with 50ml, followed by 25mls, and 10 mls
with two washes done for each volume. Two more washes were done with ice-cold 10% glycerol
at 10ml and 5ml. Finally, the cell pellet was re-suspended in 2mls of 10% glycerol. Aliquots
were made and stored at -80°C. When cells were thawed prior to use, they were pelleted and re-
suspended in 0.5M sucrose and 10% glycerol. 60µl of cells was mixed with 4ug of plasmid DNA
and left for 15 min at room temperature. The suspension was transferred to a 0.2cm gap
electroporation cuvette (BioRad, Hercules, CA) and placed in the electroporator (BioRad,
72
Hercules, CA) and pulsed with the S. aureus setting. Suspensions were immediately transferred
into 1ml of SMMP (Teknova, Hollister, CA) with 10ug/ml chloramphenicol, and incubated at
37°C at 250rpm for at least 2 hours before plating on Luria Bertani chloramphenicol plates
(10ug/ml).
Plasmid Isolation. Verification of ligation of the psmα promoter and pAMIlux was done by
testing for luminescence and sequencing the plasmid after plasmid extraction. Plasmid extraction
was carried out by growing a 5ml TSB culture with 10mg/ml chloramphenicol for selection. The
overnight cultures were used to inoculate 500-mls TSB (10mg/ml chloramphenicol) of bacterial
culture which were centrifuged for 30 minutes at 4,000g at 4°C after 24h. Cell pellets were
washed in 20mls of water and re-spun 30 minutes at 4,000g. The cells were then resuspend in
1ml of 50 mM Tris-HCl-10mM EDTA (pH8.0). 2mls of an enzyme mixture (lysozyme
(50mg/ml), lysostaphin (0.3mg/ml)) was added to the suspended cells and incubated at 37°C at
250rpm for 30 minutes. From this point on plasmid isolation was conducted with the Qiagen-tip
100 mini column per manufacturer’s protocol.
MIC Determination and Collection of Supernatants. A modified version of the CLSI
macrobroth dilution (CLSI, 2006) was used to determine linezolid, tigecycline, and clindamycin
MICs against study isolates. Tryptic Soy Broth (TSB) was used in place of Mueller-Hinton II
broth (MHB) to ensure adequate PSM production (Wang et al., 2007). Briefly, bacteria were
added to each test tube containing 2-fold drug dilutions to make a final inoculum of
approximately 1 x 10
5
CFU/ml in 4ml of TSB. Cultures were then incubated overnight at 37 °C
with shaking at 250 rpm. Additional 2-fold dilutions were made between the standard dilution
73
series in an effort to identify a more precise MIC. The MIC was read as the first tube containing
no visible growth. Additionally, MICs were determined using MHB for comparison to those
using TSB. MICs were the same for the majority of isolates tested comparing TSB and MHB;
those that differed were within one-fold dilution.
Supernatants were then collected from culture tubes containing no antibiotic, ⅛ MIC, ¼
MIC, and ½ MIC of linezolid, tigecycline, tedizolid, and clindamycin, respectively, to be used
for LC/MS/MS and immunoblotting experiments. These concentrations were chosen to reflect
clinical scenarios where sub-therapeutic concentrations may be encountered at the site of
infection due to inadequate dosing or poor drug distribution. Supernatants were stored at -80°C
with addition of 2 mg/ml ascorbic acid to minimize oxidative degradation of proteins until
assayed by LC/MS/MS or immunoblotting. Colony forming units (CFUs) for each sample at the
time of harvest of supernatants were estimated by plate counting or spectrophotometer at
OD
600nm
and compared to a standard curve of CFU vs OD
600nm
. A subset of isolates was done in
duplicate to ensure reproducibility of antibiotic effects on exoprotein production.
Quantitative LC-MS/MS. Was performed as described in chapter 3.
Western Blotting. Was performed as described in chapter 3.
Etest. Was performed as described in chapter 2.
RT-PCR. The bacterial cells from each sample were pelleted by centrifugation. RNA extraction
was performed with the Purezol RNA extraction solution (Bio-Rad, Hercules, CA). This was
74
followed by purification using the RNeasy Plus kit (Qiagen, Valencia, CA) to ensure clean
products downstream. Purified RNA was then reverse transcribed into cDNA using the qScript
super mix cDNA synthesis kit (Quanta Biosciences, Gaithersburg, MD) per manufacturer’s
conditions. Primer sequences use can be found in Table 1. The thermal cycling program
consisted of 10 min on a spectrofluorometric thermal cycler (iQ5; Bio-Rad) at 95°C, followed by
45 cycles of 30 s at 95°C, 30 s at 50°C, and 30 s at 72°C.
Table 1. Primers used for RT-PCR and cloning.
Gene/target Primer sequence (Forward/Reverse)
gyrB (F) – CAAATGATCACAGCATTTGGTACAG
(R) – CGGCATCAGTCATAATGACGAT
psmα (F) – TATCAAAAGCTTAATCGAACAATTC
(R) – CCCCTTCAAATAAGATGTTCATATC
hla (F) – AGAAAATGGCATGCACAAAAA
(R) – TATCAGTTGGGCTCTCTAAAA
spa (F) –CAGCAAACCATGCAGATGCTA
(R) – GCTAATGATAATCCACCAAATACAGTTG
RNAIII (F) – ATAGCACTGAGTCCAAGGAAACTAACT
(R) – GCCATCCCAACTTAATAACCATGT
agrA (F) – CGTAAGCATGACCCAGTTGGT
(R) – CCATCGCTGCAACTTTGTAGAC
psmα promoter (F) - AGAATTCGCATGCCTAACGTGTTATTCGTTTTAAACTTAT
(R) - GGATCCTCTAGATTTGCTTATGAGTTAACTTCATTGTA
Cytotoxicity Assays. Cytotoxicity of neutrophils was determined by an LDH release assay
(Promega, Madison, WI). Neutrophils (PMNs) were plated at 10
6
per well in a white 96-well
plate in RPMI/H. Bacteria were added at a concentration of 10
5
CFU per well. Supernatant
containing antibiotics were used at concentrations to give ½, ¼, and ⅛ MIC after adding 100ul of
bacteria suspended in RPMI/H to 100ul of PMN suspension. No antibiotic controls were
included.
Cytotoxicity against A549 lung epithelial cells was performed as described previously
(Wardenburg et al., 2007). Briefly, S. aureus strains were grown overnight in TSB at 37°C at
75
250rpm, then re-inoculated and grown to mid-exponential log phase (OD
600
= 0.4). Cells were
then pelleted and washed once with PBS before re-suspending the pellet in 10 ml F12K media.
100µl of bacteria in F12K were then added to a 96-well plate that was seeded overnight (18h)
with 15,000 A549 cells per well.
In some experiments luminescence was monitored every hour for 4 hours as well as
aliquots of supernatants taken for LDH release, generating an hourly time course of
luminescence and cytotoxicity. In other instances luminescence was monitored hourly but LDH
release was only measured after the 4h incubation using an LDH assay kit (Promega, Madison,
WI).
Statistical Analysis. Continuous variables were compared by student t-test or Wilcoxan Rank
Sum test. PSMα concentrations under various test conditions were compared by ANOVA with
Dunns correction using GraphPad Prism version 5.0 (www.GraphPad.com, San Diego, CA).
Results.
Effects of sub-inhibitory concentrations of antibiotics on growth. We first performed growth
curves on the LAC reference strain and two clinical isolates to determine cell density by OD
600nm
measurement in the presence of TR-700, linezolid, tigecycline, and clindamycin at ½, ¼, and ⅛
MIC and without antibiotic. At ½ MIC for all four agents, growth was affected at 24 h after
incubation in some cases by as much as 50% final OD compared to no antibiotic control.
Minimal to no effect on growth was observed at ¼ and ⅛ MICs for any of the agents (Figure 1).
Growth of clinical strains corresponded to that observed with USA300 control strain as indicated
by optical density values (OD
600nm
) measured by spectrophotometer at time of supernatant
76
harvest. Concentration values of PSMα peptides measured for each sample were normalized to
CFU to account for differences in growth and cell counts.
Figure 1. Growth of control strain LAC under antibiotic sub-inhibitory concentrations of
antibiotics.
0 5 10 15 20 25
0
1
2
3
4
Control
1/2 MIC
1/4 MIC
1/8 MIC
LAC TR-700
Time (h)
OD 600nm
LAC linezolid
0 5 10 15 20 25
0
1
2
3
4
Control
LZ 1/2 MIC
LZ 1/4 MIC
LZ 1/8 MIC
Time (h)
OD 600nm
LAC tigecycline
0 5 10 15 20 25
0
1
2
3
4
Control
TYG 1/2
TYG 1/4
TYG 1/8
Time (h)
OD 600nm
LAC clindamycin
0 5 10 15 20 25
0
1
2
3
4
Control
CL 1/2 MIC
CL 1/4 MIC
CL 1/8 MIC
Time (h)
OD 600nm
In some strains ½ MIC of the antibiotics used stunted growth. Lower concentrations rarely affected
growth.
Effects of TR-700 on PSMα production MRSA strains causing SSTI. Clinical strains that caused
cSSTIs were grouped by baseline PSMα production. Representative high, medium, and low
producers were selected to examine the effect of TR-700 and clindamycin on PSMα production.
Overall, 24 clinical isolates and the LAC control strain were tested. TR-700 at ½ MIC had a
pronounced inhibitory effect on PSMα production among clinical isolates, though the effect
varied for all four alpha subtypes (Figure 2).
77
Figure 2. Effects of TR-700 on PSMα
1-4
production.
Effect of Sub-inhibitory concentrations of TR-700 on PSMα production. Data represent median with IQR.
1-way ANOVA with Dunnett’s post test was used for statistical analysis.
***
= p < 0.001,
*
= p < 0.05
PSMα
3
was most inhibited in which nearly all isolates tested produced no measurable
amounts while PSMα
4
was least inhibited with production decreased to a median of 21% of
baseline value. Interestingly, paradoxical effects on PSMα production were observed for TR-700
at ¼ and ⅛ MICs with significant strain variability. PSMα production was inhibited in a dose-
dependent manner in some strains while enhanced by nearly 200% from baseline in others.
Similarly, in the LAC control strain, TR-700 at ½ MIC significantly inhibited PSMα production
to 30% of baseline, while ¼ and ⅛ MICs increased PSM production by 20% and 33%,
respectively. Of note, the percentage change at each sub-MIC above reflects the mean values of
all four PSMα alpha peptides.
78
To further investigate the paradoxical effects observed on PSMα production at low sub-
inhibitory concentrations of study drugs, we measured transcript levels of agrA and RNAIII of
the agr-quorum sensing system involved in PSMα regulation by quantitative RT-PCR in selected
strains.
Figure 3. Effects of TR-700 on agrA and RNAIII
Figure 3. Effects of TR-700 on global regulators. mRNA was extracted at 24h and expression was
normalized to housekeeping gene gyrB.
The LAC control strain plus two clinical strains that showed an increase, decrease, and minimal
change in PSMα production from baseline in the presence of ¼ and ⅛ MICs of TR-700
(tedizolid) were selected for mRNA expression analysis. Bacterial cells collected from
supernatants for PSMα analysis were used to extract mRNA to measure expression of RNAIII
and agrA. In all instances, Tedizolid induced expression of RNAIII and agrA above baseline in a
dose-dependent manner (Figure 3). Interestingly, as the sub-inhibitory concentrations of
antibiotics increased, the expression levels of the global regulators also increased despite a
general trend of decreased PSMα production, however this may be due to time of collection
(24h) in which case effects at lower concentrations may have already diminished (Yamaki et al.,
2011b).
0
1
2
3
control
"1/2"
"1/4"
"1/8"
Normalized
fold
expression
TR-‐700
concentraeon
TR-‐700
AgrA
RNA
III
79
Effects of common protein synthesis inhibiting antibiotics on PSMα production. After exploring
the effects of the second-generation oxazolididnone, tedizolid, on PSMα peptide production and
finding paradoxical increases in their production we sought to explore the effects of currently
available antibiotics. Linezolid a first-generation oxazolidinone was selected as it is in the same
family as tedizolid and is commonly used to treat MRSA infections. Little data is currently
available on tigecycline, a protein synthesis inhibitor, and its effects on any type of exoprotein
production in bacteria. As it is currently used for treatment of SSTI caused by MRSA we
included it in our study. Finally, clindamycin was included as it is the gold standard protein
synthesis inhibitor commonly used in exoprotein production inhibition studies.
To examine the effects of linezolid, tigecycline, and clindamycin at sub-inhibitory
concentrations (sub-MIC) on PSMα production, we selected 11 invasive clinical isolates and two
MRSA control strains from the isolates originally tested at baseline (chapter 3), to represent high,
medium, and low baseline PSMα producers for testing (Table 2). Bacterial isolates were grown
in the presence of each antibiotic at ½, ¼, and ⅛ MIC; supernatant was harvested after 24 h and
analyzed by LC/MS/MS.
Table 2. Baseline production of PSMα peptides by representative isolates selected for
testing with antibiotics.
PSMα
1
grouped by
baseline
production level of
PSMα
3
(µg/ml)
No. of
isolates
PSMα
1
µg/ml
mean (std)
PSMα
2
µg/ml
mean (std)
PSMα
3
µg/ml
mean (std)
PSMα
4
µg/ml
mean (std)
Very low
producers: <1
3 0.11 (0.2) 0.56 (0.62) 0.06 (0.11) 0.22 (0.29)
Low producers: 1-5 4 6.97 (3.75) 9.42 (3.15) 4.18 (1.44) 5.63 (1.16)
Medium
producers
L
: 6-15
5 11.04 (6.9) 13.51 (5.2) 7.66 (2.85) 7.95 (3.68)
High producers:
>15
1 31.2 43.4 17.8 15.5
80
Overall, linezolid exerted an inhibitory effect, with ½ MIC having the greatest effect
followed by ¼ MIC and ⅛ MIC on PSMα
1-4
production. Clindamycin followed a similar pattern
to linezolid (Figure 4).
Figure 4. Overall effects of sub-inhibitory concentrations of antibiotics on PSMα
1-4
production.
a)
a) LZ = linezolid, b) CL = clindamycin, c) TYG = tigecycline. α1, α2, α3, α4 represent PSMα
1-4
respectively. All measured PSM values (µg/ml) were normalized to OD 600nm and then calculated as
percent relative to baseline. Isolates included (n = 10) only those that produced measureable PSMα
peptides at baseline. Treatment groups were compared by ANOVA with Dunns posttest.
(*** = p < 0.001, ** = p < 0.01, * = p < 0.05)
However, at lower concentrations of ⅛ MIC, clindamycin paradoxically increased PSMα
2,4
, but
not PSMα
3
. Tigecycline had the least inhibitory effect on PSMα’s, even at concentrations of ½
MIC where linezolid and clindamycin decreased production by approximately 90%. Tigecycline
had essentially no effect and even increased PSMα production in some isolates. In fact, increased
production of all four PSMα peptides was the primary response observed in the presence of
tigecycline at the lower concentrations of ¼ and ⅛ MICs, with tigecycline as the only drug
among the three tested to increase production even at ½ MIC in three isolates. Taken together, a
drug-specific effect on PSMα production was observed when isolates are exposed to sub-
inhibitory concentrations of the protein synthesis inhibitors tested, where tigecycline at various
sub-inhibitory concentrations increased PSM production, while this only occurred with
clindamycin at ⅛ MIC. Notably, when compared to clindamycin at ⅛ MIC which significantly
81
induced PSM production in six isolates, linezolid at ⅛ MIC modestly induced PSM production in
only one isolate.
Strain-specific responses to linezolid, tigecycline, and clindamycin exposure in PSMα
production. While linezolid appears to have the greatest and tigecycline the least inhibitory
potential on PSMα production overall, there appears to be strain-specific responses to the
presence of sub-inhibitory antibiotic concentrations in PSMα production. Drug concentrations
that induced production in some isolates did not do so in others, independent of their baseline
production level. Since antibiotics exerted the most pronounced effect on PSMα
1
production,
subsequent discussions will focus on PSMα
1
as the reference peptide.
Specifically with tigecycline at ⅛ MIC, PSMα
1
production increased by greater than
150% in seven isolates and by nearly 3-fold in one isolate; induction did not occur in two isolates
at all concentrations tested (Figure 5).
Figure 5. Strain-specific effects observed with tigecycline at ⅛ MIC on PSMα
1
production
among 11 of clinical and two control isolates tested.
Strains are grouped according to baseline PSMα
1
production as very low (<1 µg/ml), low (1-5
µg/ml), medium (6-15 µg/ml), and high (>15 µg/ml) producers.
In four of these isolates we found that PSMα
1
production was induced to greater than 10 µg/ml,
which has been shown to cause significant PMN lysis. Similar results were observed with
1/8 TYG
very low
low
medium
high
0
100
200
300
400
500
% relative to baseline
82
clindamycin at ⅛ MIC for the same seven isolates which were induced by tigecycline above.
Linezolid at the same concentration resulted in increases of PSMα
1
in only three of the thirteen
isolates tested, and was actually inhibitory in five isolates and had no significant effects in the
remaining isolates. Of additional interest is the observation that two of the three non-producers at
baseline produced the peptides in the presence of sub-inhibitory concentrations of both linezolid
and tigecycline. However, it is noteworthy that PSMα production that was induced in those two
isolates did not exceed 4µg/ml, and thus would not be expected to have significant impact on
host PMNs.
Effects of linezolid, tigecycline, and clindamycin on α–hemolysin and protein A production.
Next, we sought to determine if the antibiotic effects were specific to PSMα’s or if the
paradoxical effects also occur in other exoproteins. In contrast to results observed with the
PSMα
1-4
peptides, we found that Hla and Spa production did not increase under any condition
and was either unchanged or inhibited at the tested concentrations of antibiotics regardless of
whether PSMα peptides were induced or suppressed in the isolates (Table 3).
Table 3. Toxin-specific responses to tigecycline exposure at ⅛ MIC in representative strains
Isolate Baseline PSMα
1
production (ug/ml)
Toxin Production Under Tigecycline Exposure at ⅛
MIC
% relative to
baseline
PSMα
1
% relative to
baseline
Hla
% relative to
baseline
Spa
LAC 12.8 150 98 97
Isolate 10 12.3 95 26 96
Isolate 8 10.7 145 36 90
Isolate 14 3.3 223 53 98
Note: Strains were chosen to represent different PSMα responses to tigecycline for comparison.
83
A dose-dependent suppression of Hla and Spa production was observed with the three antibiotics
tested (Figures 6 & 7). All three antibiotics inhibited Hla production at all sub-MICs tested in a
dose-dependent manner. Of the antibiotics tested, clindamycin exhibited the greatest inhibitory
activity. Spa was not inhibited by more than 15% in any of the isolates tested at ⅛ MIC for all
tested antibiotics, while ¼ MIC caused variable reduction (15 - 95%) from baseline and ½ MIC
inhibited production by at least 90% in every isolate.
84
Figure 6. Effects of linezolid, clindamycin, and tigecycline on Hla production.
a)
No LZ ⅛ LZ ¼ LZ ½ LZ No CL ½ CL ¼ CL ⅛ CL No TYG ½TYG ¼ TYG ⅛ TYG
b)
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
linezolid
% relative to baseline
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
clindamycin
% relative to baseline
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
tigecycline
% relative to baseline
Western blot results for Hla under linezolid treatment for a representative strain (a). Combined data from
all isolates under stated condition (n=9) for linezolid, clindamycin, and tigecycline (b). Values were
calculated based on densitometry pixels normalized to OD and then calculated as a percentage of control.
Treatment groups were compared by ANOVA with Dunns posttest. * p <0.05, ** p = < 0.01, *** p < 0.001
Figure 7. Effects of linezolid, clindamycin, and tigecycline on Spa production.
a)
No abx ⅛ LZ ¼ LZ ½ LZ No abx ⅛ CL ¼ CL ½ CL ⅛ TYG ¼ TYG ½ TYG
b)
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
80
100
linezolid
% relative to baseline
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
80
100
clindamycin
% relative to baseline
1/8 MIC
1/4 MIC
1/2 MIC
0
20
40
60
80
100
tigecycline
% relative to baseline
a) Western blot results on Spa production under linezolid treatment for a representative strain. b)
Combined data from all isolates under stated condition (n=4) for linezolid, clindamycin, and tigecycline.
Values were calculated based on densitometry pixels normalized to OD and then calculated as a percentage
of control. Treatment groups were compared by ANOVA with Dunns posttest. * p <0.05, ** p = < 0.01
***
*
**
***
*
**
**
**
*
85
Tigecycline and linezolid effects on psmα, hla, agrA, and RNAIII mRNA expression. We next
sought to explore if global regulators known to control the expression of psmα were also affected
under these conditions. RNAIII is the effector molecule of the agr system that has been
demonstrated to be involved the regulation of psmα and a number of exoproteins including hla
and spa (Wang et al., 2007). Uniquely, psmα expression has recently been shown to be up-
regulated by direct binding of the AgrA peptide to the promoter region.
Thus, we used RT-PCR to quantitate mRNA expression under ¼ and ⅛ MIC conditions
of tigecycline and linezolid with the LAC control strain and strain #8, which had the highest
level of PSMα induction. The timing of optimal psmα expression without the addition of
antibiotics was found to be ~16h, therefore bacterial cells were collected after 16h of growing in
the presence of sub-MIC antibiotics. We found that under ¼ and ⅛ MIC of tigecycline, psmα
mRNA expression was highly induced compared to non-treatment controls in both test strains
with strain #8 expression being higher than LAC. At ¼ MIC of tigecycline LAC psmα
expression was induced 4-fold relative to control while strain #8 was induced to 5-fold. At ⅛
MIC in the LAC strain psmα expression was induced only 3-fold and strain #8 remained at near
5-fold (Figure 8).
86
Figure 8. Effects of linezolid and tigecycline on exotoxin mRNA levels.
a) LAC b) Strain 8
psm
hla
Spa
0
2
4
6
8
Control
1/4 LZ
1/4 TYG
1/8 LZ
1/8 TYG
Relative fold expression
RT-PCR was performed on the LAC control and clinical strain #8 with ¼ and ⅛ MIC linezolid (LZ) and
tigecycline (TYG). psmα, hla, and spa transcripts were normalized to gyrB. psmα promoter was increased
at least 3 fold under ¼ and ⅛ MIC TYG in LAC (a), while Strain #8 was induced at least 5 fold under
these conditions (b). hla and spa were inhibited at both concentrations of both drugs.
These results are similar to results found with the PSMα peptides, where strain #8 PSMα
production was induced more than in LAC. Linezolid treatment resulted in minimal induction at
¼ MIC, resulting in a 2-fold increase in strain #8 and not in the LAC strain, which is also
consistent with what was observed with PSMα peptides. Interestingly, despite the induction of
psmα expression under antibiotic conditions, hla expression was not increased in either of the
two strains and was actually decreased under both linezolid and tigecycline concentrations,
similar to results obtained from immunoblot analysis of Hla.
As an additional method for confirming increases in psmα activity, we used the pAMIlux
plasmid that we constructed to have the lux operon under the control of the psmα promoter. A
bacterial lawn was plated with both LAC and isolate #8 both of which were transformed with the
psmα pAMIlux plasmid and an eTest strip of tigecycline and linezolid was placed and
bioluminescence was measured after 12h of growth. We found a significant increase in psmα
promoter activity within the ellipse formed around the eTest strip, with the strongest signal
coming from bacteria closest to the strip. The increase in luminescence around from the bacteria
psm
hla
Spa
0
2
4
6
8
Control
1/4 LZ
1/4 TYG
1/8 LZ
1/8 TYG
Relative fold expression
87
around the eclipse is indicative of increases activity of the psmα promoter at sub-inhibitory
concentrations. Interestingly, linezolid produced stronger activity compared to tigecycline in
both strains, which may be the result of tigecycline oxidative degradation that has been reported
if media plates have been exposed to air and not immediately used (Bradford et al., 2005) (Figure
9).
Figure 9. psmα promoter activity as measure by luminescence.
a) Strain #8 linezolid b) Strain #8 tigecycline
c) LAC linezolid b) LAC tigecycline
psmα promoter activity as measured by bioluminescence. a & b demonstrate strain #8 under linezolid and
tigecycline, respectively. c & d represent LAC under linezolid and tigecycline, respectively. The point where
the ellipse intersects the eTest strip is considered the MIC and any concentrations immediately outside the
elliptical zone of clearance are sub-inhibitory.
88
When examining the global regulators agrA and RNAIII we found that they were not as
significantly affected compared to psmα expression. There was no increase expression of either
global regulator mRNA with LAC under treatment both concentrations of linezolid and
tigecycline. However, RNAIII under treatment with ¼ MIC of tigecycline did result in slight
induction (2-fold) in strain #8 (Figure 10).
Figure 10. Changes in global regulators of exotoxin production under antibiotic pressure.
a) b)
LAC
RNAIII
agrA
0.0
0.5
1.0
1.5
2.0
2.5
Control
1/4 LZ
1/4 TYG
1/8 LZ
1/8 TYG
Relative fold expression
Strain #8
RNAIII
agrA
0.0
0.5
1.0
1.5
2.0
2.5
Control
1/4 LZ
1/4 TYG
1/8 LZ
1/8 TYG
Relative fold expression
RNAIII and agrA which have both been shown to control psm expression were measured by RT-PCR.
Target gene expression was normalized to gyrB. a) LAC global regulators were decreased under tested
conditions. However, RNAIII was increased in strain #8 under tigecycline conditions (b).
Effects of tigecycline on cytotoxicity of A549 and human neutrophils. As we had previously
shown that PSMα peptides were increased, but Hla and Spa production was decreased under
antibiotic treatment. We determined whether antibiotics confer protection against toxin-mediated
effects in target host cells, human PMNs and lung epithelial cells. Using tigecycline as it had the
greatest induction, we incubated both the LAC and strain #8 with human neutrophils (PMNs)
and A549 cells in media containing no antibiotic, ½, ¼, and ⅛ MIC of tigecycline and measured
LDH as marker for cytotoxicity.
Cytotoxicity of A549 lung epithelial cells was approximately 30% when incubated with
the LAC strain after 4h. Under treatment with sub-MIC tigecycline, cytotoxicity was decreased
89
significantly at all tested concentrations (p < 0.0001) (Figure 11a). Controls of bacteria in media
containing antibiotic were included to account for any effects on bacterial growth and no growth
inhibition was observed under treatment conditions using OD
600nm
. Strain #8 was highly
cytotoxic compared to the LAC strain, cytotoxicity of A549 cells was approximately 72% after
4h incubation. However, unlike the LAC control strain, at sub-inhibitory concentrations of
tigecycline there was no significant decrease in cytotoxicity by strain #8 (Figure 11b).
Interestingly, psmα promoter activity as measured by bioluminescence each hour during the 4h
incubation period, increased over the first two to three hours, as depicted in figures 11. psmα
promoter activity under all sub-MIC conditions was higher compared to the no antibiotic control,
similar to results obtained from RT-PCR. This suggests that the psmα promoter was active and
most likely PSMα peptides were being produced, however this did not lead to increased
cytotoxicity. Cytotoxicity of LAC and strain #8 were similar against PMNs (~50%). No
significant changes in PMN cytotoxicity were noted under tigecycline conditions with isolate #8
and LAC (Figure 12a & b). A deletion mutant of LAC was also included in experiments to
determine the impact of PSMαs on PMN cytotoxicity. There was an overall decrease in PMN
cytotoxicity by approximately 50% with the LACΔpsm mutant. Monitoring of the psmα
promoter showed there was increased activity but there was no differences in treatment vs
control, as seen with LAC wt (Figure 12c).
90
Figure 11. Cytotoxicity against A549 cells under sub-MIC tigecycline.
a) LAC b) Strain 8
0
10
20
30
40
0.0
0.5
1.0
1.5
2.0
2.5
Control
1/2 TYG
1/4 TYG
1/8 TYG
Control
1/2 TYG
1/4 TYG
1/8 TYG
****
% A549 cytotoxicity
LU/OD
0
20
40
60
80
100
0.0
0.5
1.0
1.5
2.0
2.5
Control
1/2 TYG
1/4 TYG
1/8 TYG
Control
1/2 TYG
1/4 TYG
1/8 TYG
% A549 cytotoxicity
LU/OD
Cytotoxicity against A549 cells is displayed on the left side of the graph matched with luminescence of the
psmα promoter on the right side after a 4h co-incubation. Lysis of A549 was significantly decreased with
the LAC control strain under all sub-MIC TYG (a), cytotoxicity of strain #8 was not significantly
decreased at sub-MIC TYG (b) although not as strongly. (**** p < 0.0001)
Figure 12. Cytotoxicity against human PMNs under sub-MIC tigecycline.
a) LAC b) Strain 8
c) LAC Δpsm
0
10
20
30
40
0.0
0.5
1.0
1.5
2.0
Control
1/2 TYG
1/4 TYG
1/8 TYG
Control
1/2 TYG
1/4 TYG
1/8 TYG
% PMN cytotoxicity
LU/OD
No significant decreases or increases were observed with the LAC control strain (a) or strain #8 (b) in
cytotoxicity against human PMNs. There was a trend for decreases at ½ MIC for both strains but no
significant changes in cytotoxicity at ¼ MIC and ⅛ MIC. The LAC psmα deletion mutant was included to
determine the impact of PSMαs on PMN cytotoxicity while monitoring psm promoter activity under
tigecycline conditions (c).
0
20
40
60
80
0.0
0.5
1.0
1.5
2.0
2.5
Control
1/2 TYG
1/4 TYG
1/8 TYG
Control
1/2 TYG
1/4 TYG
1/8 TYG
% PMN cytotoxicity
LU/OD
0
20
40
60
80
0
1
2
3
4
% PMN cytotoxicity
LU/OD
Control
1/2 TYG
1/4 TYG
1/8 TYG
Control
1/2 TYG
1/4 TYG
1/8 TYG
91
Discussion.
Current literature suggests that CA-MRSA strains produce virulence factors such as alpha-
hemolysin at higher concentrations than HA-MRSA strains and those antibiotics that inhibit
protein synthesis can decrease exoprotein production at sub-inhibitory concentrations in S.
aureus. Based on these studies that demonstrated exotoxin suppression by certain protein
synthesis inhibiting antibiotics (Dumitrescu et al., 2008; Stevens et al., 2007a), clinicians may
make the generalized assumption of this antibiotic class that even at sub-therapeutic levels these
drugs can provide additional benefits.
Here we investigated the anti-virulence potential of tedizolid, linezolid, clindamycin, and
tigecycline at sub-MICs. We chose the tested concentrations at which bacterial growth would be
minimally affected and to reflect clinical scenarios where sub-MIC concentrations might be
achieved at sites of infection due to inadequate dosing or altered pharmacokinetic parameters of
the patient. We are the first to investigate the potential anti-virulence effect of tigecycline and
tedizolid on PSMα
1-4
peptide production by MRSA isolates. We also included linezolid and
clindamycin for comparison and tested the anti-virulence effects of antibiotics on two other key
exotoxins, Hla and Spa. We found that at sub-MICs, agents from this class of antibiotics have
pleiotropic effects on toxin production that is dependent on strain, drug, and toxin tested.
Overall, we found that low concentrations (¼ and ⅛ MICs) of the antibiotics tested were
more likely to induce PSMα production, particularly with tigecycline and less so with tedizolid
and clindamycin followed by linezolid. While, linezolid, tedizolid, and clindamycin at ½ MIC
inhibited PSMα production, tigecycline lead to increased production or exerted no inhibitory
effects. The increase at ½ MIC with tigecycline is particularly interesting as it demonstrates that
while growth overall maybe stunted, PSMα production can still be increased. These results
92
demonstrated that different drugs within the same pharmacological class could exert different
effects on exoprotein production from MRSA clinical isolates. Furthermore, when comparing
effects of the same drug among different clinical isolates, we found that strain responses to drug
exposure in exotoxin production vary. The magnitude of induction and inhibition differed among
isolates where the same drug could induce PSMα production in one isolate while inhibiting
production in another. Taken together, our results suggest that strain-specific and drug-specific
effects exist with exoprotein production under antibiotic exposure.
The observation of PSMα induction by protein synthesis inhibiting antibiotics at sub-
MICs (e.g. linezolid and clindamycin) is contrary to published literature on other exotoxins,
where studies have consistently found suppression under this class of antibiotics (Dumitrescu et
al., 2008; Dumitrescu et al., 2007; Ohlsen et al., 1998; Stevens et al., 2007a). To examine
whether the effects of PSMα induction are specific for these peptides, we examined production
of Hla and Spa under the same treatment conditions with tigecycline, linezoid, and clindamycin.
Interestingly we found production of both Hla and Spa to be suppressed by antibiotics tested at
sub-inhibitory concentrations among the study isolates, independent of antibiotic effect on PSMα
production in the same isolates. Hence, our results indicate that the induction effects may be
specific for PSMα peptides.
We hypothesized that the increase in PSMα production under sub-MIC conditions would
lead to increases in cytotoxicity against A549 cells and PMNs. To our surprise our in-vitro A549
cellular model did not demonstrate increases in cytotoxicity but actually decreases in
cytotoxicity, despite increased psmα promoter activity under these conditions. The likely
explanation is that while the PSMα peptides and psmα promoter activity are increased under
these conditions, the inhibition observed with Spa and Hla are at least partially responsible for
93
the decrease in cytotoxicity, particularly Hla since it has been demonstrated to be highly
cytotoxic against A549 cells. It is also likely other exotoxins responsible for S. aureus virulence
are also inhibited by sub-MIC concentrations of protein synthesis inhibitors, leading to decreases
in cytotoxicity. Interestingly we found the cytotoxicity of strain #8 against PMNs at ¼ and ⅛
MIC of tigecycline slightly increased compared to non-treatment controls, this is in contrast to
what was found with A549 cells (figures 10 & 11). This is likely due to the increase sensitivity
of PMNs to PSMα peptides compared to A549 cells.
The exact mechanism responsible for the increased production of PSMα in the presence
of sub-inhibitory concentrations of protein synthesis inhibitors is unknown. Joo et al. observed
similar finding with clindamycin and tetracycline at 1/10 MIC on PSMα production. They
suggested that agr activity is involved whereby expression of RNAIII along with psmα mRNA is
increased under antibiotic pressure, (Joo et al., 2010). In support of this notion, we also found
increased expression of RNAIII in strain #8 and increased psmα expression in both strain #8 and
LAC in the presence of sub-inhibitory concentrations of tigecycline. In addition, the differential
modulation by protein synthesis inhibitors on exotoxin production is of interest. PSMα is more
frequently induced by clindamycin, and tigecycline at sub-MIC, but is almost always inhibited
by linezolid.
One possibility for this observation is that while these antibiotics may induce a stress
response leading to increases in psmα transcription, linezolid maybe more effective at inhibiting
overall protein synthesis since it is known to inhibit initiation of protein synthesis by blocking
the formation of the 70S initiation complex, rather than only blocking the peptidyl-transferase
reaction or tRNA binding as with clindamycin and tigecycline, respectively (Olson et al., 2006;
Swaney et al., 1998; Tenson et al., 2003). It is notable that at the same concentration of protein
94
synthesis inhibitor (e.g. tigecycline), PSMα production was induced in certain isolates while
inhibited in others. We speculate that variation in virulence regulation among clinical isolates
likely account for the differences observed. Furthermore, why Hla and Spa were inhibited in all
cases despite increase in PSMα production is unknown, but may be due to the unique positive
regulatory pathway involving AgrA binding to the psmα promoter but not hla or spa promoters.
Our study has several limitations. First, our observations are strictly in-vitro and may not reflect
how MRSA strains express virulence in the host environment where the effects of toxin
induction may be abrogated by host defenses that are present. Additionally, we found that ½
MIC of the tested drugs could affect growth or final cell counts in some isolates. We have
accounted for changes in cell counts by normalizing the amount of toxin measured to CFU.
However, other factors such as initial slowing in growth, while not affecting final CFU count
could also affect toxin production. Notably, even with potential inhibition on bacterial growth for
tigecycline at ½ MIC, lower PSMα
1-4
production was not observed.
Further investigation is needed to examine if any other major toxins are induced, as are
PSMα peptides when exposed to sub-inhibitory concentrations of different antibiotics that act by
inhibiting protein synthesis. Additionally, elucidating the mechanism that underlies induction of
PSMα peptides is of great interest and has implications on the future development of new
antibiotics. The clinical relevance of toxin induction and inhibition on outcomes of invasive
infections deserves further study based on our in vitro observations. For now, our data caution
against the general assumption by clinicians that all protein synthesis inhibitors possess
inhibitory potential against all exotoxins produced by S. aureus. Our data further affirms the
importance of adequate dosing when administering these agents for the treatment of invasive
95
infections to minimize the potential adverse effect associated with sub-MIC exposure resulting in
exoprotein induction.
This chapter was based on the following publications:
1. Yamaki J, Synold T, Wong-Beringer A. Antivirulence potential of TR-700 and clindamycin
on clinical isolates of Staphylococcus aureus producing phenol-soluble modulins. Antimicrob
Agents Chemother. 2011 Sep;55(9):4432-5
2. Yamaki J, Synold T, Wong-Beringer A. Virulence factor inhibition by linezolid, tigecycline, and
clindamycin against isolates causing invasive infections. Antimicrob Agents Chemother 2013
(Conditionally accepted April 2013)
96
Chapter 5.
Summary and Future Directions
MRSA has become a leading pathogen in both the community and hospital settings
leading to significant morbidity and mortality. Since its emergence in the late 1990’s,
community-associated MRSA strains have become frequently responsible for SSTIs as well as
severe diseases such as necrotizing pneumonia. The success as a pathogen and severity of
disease it causes is thought to be related to a number of virulence factors it produces. The data
presented in this thesis demonstrated a change in the epidemiology of MRSA strains within an
institution in Los Angeles County, identified novel markers that can be used to guide therapy,
explored the contribution of in-vitro toxin production and resistance to patient outcomes and
disease severity, and identified a phenomenon of exotoxin induction by protein synthesis
inhibiting antibiotics.
Identifying patients at risk for severe infection or clinical failure is key to administering
proper antibiotics and avoiding unnecessary treatments. We documented a shift in the
epidemiology of MRSA infections, where CA-MRSA strains have become responsible for a
significant portion of MRSA infections within the healthcare setting. That study was based on
the premise that CA-MRSA strains have been described as being more susceptible to non-β-
lactam antibiotics since these strains are less likely to have encountered selective pressures
within the healthcare settings. Since we documented a shift in CA-MRSA epidemiology we
wanted to assess if the commonly described CA-MRSA phenotype of increased susceptibility
was still true for vancomycin. Using molecular biology techniques to type MRSA strains we
have found that in strains collected from 2005 to 2007, susceptibility to vancomycin could be
predicted by possession of the genes encoding PVL, thus with the development and use of rapid
97
diagnostics for identifying these genes, empiric vancomycin and alternative therapies can
possibly be guided by this marker. Extending the time period from 2007 all the way to 2011, we
found that the predictability of vancomycin susceptibility was still true in pneumonia cases,
however not for bacteremia.
In the previous study bacterial markers were shown to be a possible means for guiding
MRSA therapy. We have also provided a timeframe that can be used as a window of opportunity
to assess patient therapy when administering vancomycin for bacteremia treatment. Response
after receipt of vancomycin for 72h was the strongest predictor of success or failure. Patients
who responded to vancomycin therapy at this point were more likely to be successfully treated
by the end of therapy, where as those who did not respond but continued to receive vancomycin
despite lack of response were most likely to fail treatment overall. Those who were switched to
an alternative agent at 72h because of failure to respond had better odds of success than those
who were not switched. Thus, providing a marker from the patient perspective in this case as
opposed to the bacterial side as previously described.
Resistance has been suggested to contribute to MRSA morbidity and mortality, especially
in bacteremia, as mentioned above. Contrary to other investigators we did not find any
association between elevated MICs for vancomycin and poor clinical outcomes. There was only
a trend in correlating vancomycin MIC >1µg/ml with poor outcomes (p=0.06), where other
studies have found much stronger significance in similar analysis. The reason for the lack in
significance between the two variables could be attributed to a number of reasons, including
differences in patient population, as our population consists mostly of elderly, other confounding
variables that may have contributed but were not part of the multi-variate analysis, and finally
the size of our cohort may be inadequately powered to detect a significant difference.
98
Resistance was not found to contribute significantly to outcomes in our cohort of patients,
leaving the possibility of virulence contributing to outcomes. Other investigators have
demonstrated that certain virulence factors produced by MRSA strains contribute to pathogenesis
and disease severity in animal models. Furthermore, laboratory CA-MRSA strains have been
shown to have more active agr systems compared to HA-MRSA strains that would be expected
to lead to higher amounts of exotoxin production between the two. In-vivo animal models have
been used to determine exotoxin contribution to disease in MRSA infections and based on other
bacterial diseases such as C. difficile and E. coli O157:H7 where in-vitro exotoxin production
was shown to correlate with both animal models and human infection, we sought to test if the
same is true in MRSA clinical strains that cause a wide spectrum of infections. No correlation
between in-vitro Hla, Spa, or PSMα production and disease severity was found. Nor was any
correlation between cytotoxicity of A549 and PMNs with disease severity. Interestingly, we did
find differences in susceptibility of A549 and PMNs depending on MRSA strain background,
where A549 cells infected with HA-MRSA strains were lysed more effectively than PMNs with
the same HA strains, and CA-MRSA strains were more cytotoxic against PMNs compared to
A549 cells.
In clinical practice clinicians often prescribe protein synthesis inhibiting antibiotics with
the expectation that exoprotein production would be inhibited. For example, it is common
practice in the treatment of necrotizing fasciitis caused by Streptococcus pyogenes to add
clindamycin to the treatment regimen, as this disease is known to be toxin mediated. With
emerging animal data indicating MRSA infections are toxin mediated and additional studies
demonstrating exotoxin inhibition in MRSA, clinicians may be more inclined to prescribe these
antibiotics, since even at sub-inhibitory concentrations protein synthesis inhibitors have been
99
shown to inhibit toxin production. With emerging data demonstrating the role of PSMs and Hla
in infections such as pneumonia and skin soft tissue infections, we sought to examine the effects
of sub-inhibitory concentrations of commonly prescribed protein synthesis inhibiting antibiotics
on PSMα production, as no studies had yet been conducted on this topic. We found that protein
synthesis inhibitors can inhibit exoprotein production as expected. However to our surprise we
found that paradoxical increases in PSMα peptides can occur. These effects were drug specific
with tigecycline leading to increasing in production more frequently than any other drug test,
while linezolid was the least likely to cause induction. Strain specific responses were also noted
as certain isolates responded differently than others under the same treatment conditions. Most
importantly, the effects of paradoxical increases appeared to be specific to PSMα peptides as Hla
and Spa were inhibited in all cases under all conditions.
Overall, this thesis was highly focused on improving treatment for patients with MRSA
infections. Our studies have focused on identifying patient and bacterial characteristics that could
potentially be used to guide antibiotic therapy to improve patient outcomes. Based on these
studies, practice can be potentially improved if rapid diagnostics for detecting bacterial markers
for empiric vancomycin treatment become available on market. Likewise the 72h timeframe for
evaluation of patient response to vancomycin therapy is likely to improve patient outcomes when
incorporated into clinical practice. Importantly, this can be immediately implemented in clinical
practice, although more studies need to be conducted exploring how changing to alternative
therapy may affect outcomes, the 72h mark can at least provide a point of assessment where
clinicians can use their clinical judgment with the patients’ therapy.
Furthermore, we investigated the effects of current treatments used for MRSA infections
and their effects on exoprotein production and found tigecycline has very strong potential for
100
inducing PSMα peptide production among MRSA strains subjected to sub-MIC levels. These
results indeed bring into question the general assumption that clinicians currently have regarding
the role of protein synthesis inhibitors in the treatment of toxin-mediated infections caused by S.
aureus. Clinically, this effect is most relevant when treating patients with deep-seated infections
where only a fraction of antibiotic concentrations in the blood is achieved at the site of infection.
Thus, an increase in toxin production mediated by selected antibiotics can lead to adverse
outcomes in patients. Lastly, we investigated if severity of MRSA infections like other toxin
mediated bacterial diseases can be correlated with in-vitro toxin production, but found little
correlation between the two.
Future directions.
Based on these findings additional studies are required to continue from the observations
made. Our studies were largely retrospective in nature, thus prospective studies particularly in
measuring outcomes after switching to alternative antibiotics based on response at 72h are
warranted. Evaluation of ceftaroline in the treatment of MRSA pneumonia is also of great
importance, in-vitro susceptibility of our pneumonia isolates suggest it may be a good alternative
as clinical strains that caused pneumonia remained highly susceptible to ceftaroline. In our
pneumonia patient population vancomycin was the primary drug used as empiric and directed
therapy. While we did not demonstrate a correlation with poor treatment outcomes with high
vancomycin MICs, others have found a relationship in both bacteremia and pneumonia (Haque et
al., 2010; Sakoulas et al., 2004). Based on these findings it is crucial to determine if the newest
MRSA therapy, ceftaroline, is efficacious in the treatment of MRSA pneumonia.
101
Ample evidence mentioned earlier suggests that certain exotoxins play critical roles in the
pathogenesis and severity of certain types of animal models of MRSA infections. This has been
shown to be true in a number of other bacterial infections (C. difficile) where in-vitro toxin
production can be correlated with animal models and human disease. We did not find an
association between patient outcomes and disease severity in our pneumonia strains. A number
of explanations are possible for this observation, one being the conditions used in-vitro do not
mimic conditions required for expression of virulence in-vivo during human infections. Another
important possibility is the host immune response due to patients having previously encountered
S. aureus may have developed antibodies or immune memory that can mitigate disease severity
when re-infected. Thus, the involvement of the host response in MRSA infections should be
explored in prospective clinical studies.
Finally, the finding that PSMα peptide production can be increased under sub-MIC
concentrations of protein synthesis inhibiting antibiotics is concerning. We not only
demonstrated that tigecycline is the most likely of the antibiotics tested to induce PSMα
production but that these effects can lead to increased cytotoxicity of PMNs in strains that are
highly induced. Future work in the laboratory of Dr. Wong-Beringer in collaboration with Dr.
Rodgers is planned to confirm this observation in-vivo by monitoring psm promoter activity by
bioluminescence in a murine model of SSTI with sub-MIC tigecycline and correlating with
disease severity.
102
References
(2003a). From the Centers for Disease Control and Prevention. Public health dispatch: outbreaks
of community-associated methicillin-resistant Staphylococcus aureus skin infections--Los
Angeles County, California, 2002-2003. Jama 289, 1377.
(2003b). Methicillin-resistant staphylococcus aureus infections among competitive sports
participants--Colorado, Indiana, Pennsylvania, and Los Angeles County, 2000-2003. MMWR
Morb Mortal Wkly Rep 52, 793-795.
[Anon] (2005). Guidelines for the management, of adults with hospital-acquired, ventilator-
associated, and healthcare-associated pneumonia. Am J Resp Crit Care 171, 388-416.
Abdelnour, A., Arvidson, S., Bremell, T., Ryden, C., and Tarkowski, A. (1993). The Accessory
Gene Regulator (Agr) Controls Staphylococcus-Aureus Virulence in a Murine Arthritis Model.
Infect Immun 61, 3879-3885.
Barrett, T.W., and Moran, G.J. (2004). Update on emerging infections:news from the Centers for
Disease Control and Prevention. Methicillin-resistant Staphylococcus aureus infections among
competitive sports participants--Colorado, Indiana, Pennsylvania, and Los Angeles County,
2000-2003. Ann Emerg Med 43, 43-45; discussion 45-47.
Bartlett, A.H., Foster, T.J., Hayashida, A., and Park, P.W. (2008). Alpha-toxin facilitates the
generation of CXC chemokine gradients and stimulates neutrophil homing in Staphylococcus
aureus pneumonia. J Infect Dis 198, 1529-1535.
103
Benito, Y., Kolb, F.A., Romby, P., Lina, G., Etienne, J., and Vandenesch, F. (2000). Probing the
structure of RNAIII, the Staphylococcus aureus agr regulatory RNA, and identification of the
RNA domain involved in repression of protein A expression. Rna 6, 668-679.
Bernard, G.R., Artigas, A., Brigham, K.L., Carlet, J., Falke, K., Hudson, L., Lamy, M., Legall,
J.R., Morris, A., and Spragg, R. (1994). The American-European Consensus Conference on
ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir
Crit Care Med 149, 818-824.
Bernardo, K., Pakulat, N., Fleer, S., Schnaith, A., Utermohlen, O., Krut, O., Muller, S., and
Kronke, M. (2004). Subinhibitory concentrations of linezolid reduce Staphylococcus aureus
virulence factor expression. Antimicrob Agents Ch 48, 546-555.
Bhakdi, S., Bayley, H., Valeva, A., Walev, I., Walker, B., Weller, U., Kehoe, M., and Palmer, M.
(1996). Staphylococcal alpha-toxin, streptolysin-O, and Escherichia coli hemolysin: Prototypes
of pore-forming bacterial cytolysins. Arch Microbiol 165, 73-79.
Bhakdi, S., Muhly, M., Mannhardt, U., Hugo, F., Klapettek, K., Muellereckhardt, C., and Roka,
L. (1988). Staphylococcal Alpha-Toxin Promotes Blood-Coagulation Via Attack on Human-
Platelets. Journal of Experimental Medicine 168, 527-542.
Blickwede, M., Wolz, C., Valentin-Weigand, P., and Schwarz, S. (2005). Influence of
clindamycin on the stability of coa and fnbB transcripts and adherence properties of
Staphylococcus aureus Newman. Fems Microbiol Lett 252, 73-78.
Boisset, S., Geissmann, T., Huntzinger, E., Fechter, P., Bendridi, N., Possedko, M., Chevalier,
C., Helfer, A.C., Benito, Y., Jacquier, A., et al. (2007). Staphylococcus aureus RNAIII
104
coordinately represses the synthesis of virulence factors and the transcription regulator Rot by an
antisense mechanism. Gene Dev 21, 1353-1366.
Bonnstetter, K.K., Wolter, D.J., Tenover, F.C., McDougal, L.K., and Goering, R.V. (2007).
Rapid multiplex PCR assay for identification of USA300 community-associated methicillin-
resistant Staphylococcus aureus isolates. J Clin Microbiol 45, 141-146.
Boye, K., Bartels, M.D., Andersen, I.S., Moller, J.A., and Westh, H. (2007). A new multiplex
PCR for easy screening of methicillin-resistant Staphylococcus aureus SCCmec types I-V. Clin
Microbiol Infec 13, 725-727.
Bradford, P.A., Petersen, P.J., Young, M., Jones, C.H., Tischler, M., and O'Connell, J. (2005).
Tigecycline MIC testing by broth dilution requires use of fresh medium or addition of the
biocatalytic oxygen-reducing reagent Oxyrase to standardize the test method. Antimicrob Agents
Ch 49, 3903-3909.
Brown, E.L., Dumitrescu, O., Thomas, D., Badiou, C., Koers, E.M., Choudhury, P., Vazquez, V.,
Etienne, J., Lina, G., Vandenesch, F., et al. (2009). The Panton-Valentine leukocidin vaccine
protects mice against lung and skin infections caused by Staphylococcus aureus USA300.
Clinical microbiology and infection : the official publication of the European Society of Clinical
Microbiology and Infectious Diseases 15, 156-164.
Bubeck Wardenburg, J., and Schneewind, O. (2008). Vaccine protection against Staphylococcus
aureus pneumonia. The Journal of experimental medicine 205, 287-294.
Chan, P.F., and Foster, S.J. (1998). Role of SarA in virulence determinant production and
environmental signal transduction in Staphylococcus aureus. J Bacteriol 180, 6232-6241.
105
Cheung, A.L., Bayer, A.S., Zhang, G.Y., Gresham, H., and Xiong, Y.Q. (2004). Regulation of
virulence determinants in vitro and in vivo in Staphylococcus aureus. Fems Immunol Med Mic
40, 1-9.
Cheung, A.L., Eberhardt, K.J., Chung, E., Yeaman, M.R., Sullam, P.M., Ramos, M., and Bayer,
A.S. (1994). Diminished Virulence of a Sar(-)/Agr(-) Mutant of Staphylococcus-Aureus in the
Rabbit Model of Endocarditis. J Clin Invest 94, 1815-1822.
Chua, T., Moore, C.L., Perri, M.B., Donabedian, S.M., Masch, W., Vager, D., Davis, S.L.,
Lulek, K., Zimnicki, B., and Zervos, M.J. (2008). Molecular epidemiology of methicillin-
resistant Staphylococcus aureus bloodstream isolates in urban Detroit. J Clin Microbiol 46, 2345-
2352.
CLSI (2006). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow
Aerobically; Approved Standard, Vol Seventh Edition.
David, M.Z., and Daum, R.S. (2010). Community-Associated Methicillin-Resistant
Staphylococcus aureus: Epidemiology and Clinical Consequences of an Emerging Epidemic.
Clin Microbiol Rev 23, 616-+.
Davis, S.L., Rybak, M.J., Amjad, M., Kaatz, G.W., and McKinnon, P.S. (2006). Characteristics
of patients with healthcare-associated infection due to SCCmec type IV methicillin-resistant
Staphylococcus aureus. Infect Cont Hosp Ep 27, 1025-1031.
Diep, B.A., Afasizheva, A., Le, H.N., Kajikawa, O., Matute-Bello, G., Tkaczyk, C., Sellman, B.,
Badiou, C., Lina, G., and Chambers, H.F. (2013). Effects of Linezolid on Suppressing In Vivo
106
Production of Staphylococcal Toxins and Survival Outcomes in a Rabbit Model of MRSA
Necrotizing Pneumonia. J Infect Dis.
Diep, B.A., Chan, L., Tattevin, P., Kajikawa, O., Martin, T.R., Basuino, L., Mai, T.T., Marbach,
H., Braughton, K.R., Whitney, A.R., et al. (2010). Polymorphonuclear leukocytes mediate
Staphylococcus aureus Panton-Valentine leukocidin-induced lung inflammation and injury. P
Natl Acad Sci USA 107, 5587-5592.
Diep, B.A., and Otto, M. (2008). The role of virulence determinants in community-associated
MRSA pathogenesis. Trends Microbiol 16, 361-369.
Doss, S.A., Tillotson, G.S., and Amyes, S.G.B. (1993). Effect of Sub-Inhibitory Concentrations
of Antibiotics on the Virulence of Staphylococcus-Aureus. J Appl Bacteriol 75, 123-128.
Dumitrescu, O., Badiou, C., Bes, M., Reverdy, M.E., Vandenesch, F., Etienne, J., and Lina, G.
(2008). Effect of antibiotics, alone and in combination, on Panton-Valentine leukocidin
production by a Staphylococcus aureus reference strain. Clin Microbiol Infect 14, 384-388.
Dumitrescu, O., Boisset, S., Badiou, C., Bes, M., Benito, Y., Reverdy, M.E., Vandenesch, F.,
Etienne, J., and Lina, G. (2007). Effect of antibiotics on Staphylococcus aureus producing
Panton-Valentine leukocidin. Antimicrob Agents Chemother 51, 1515-1519.
Forsgren, A., and Nordstrom, K. (1974). Protein A from Staphylococcus aureus: the biological
significance of its reaction with IgG. Ann N Y Acad Sci 236, 252-266.
Francis, J.S., Doherty, M.C., Lopatin, U., Johnston, C.P., Sinha, G., Ross, T., Cai, M., Hansel,
N.N., Perl, T., Ticehurst, J.R., et al. (2005). Severe community-onset pneumonia in healthy
107
adults caused by methicillin-resistant Staphylococcus aureus carrying the Panton-Valentine
leukocidin genes. Clin Infect Dis 40, 100-107.
Frank, K.M., Zhou, T., Moreno-Vinasco, L., Hollett, B., Garcia, J.G., and Bubeck Wardenburg,
J. (2012). Host response signature to Staphylococcus aureus alpha-hemolysin implicates
pulmonary Th17 response. Infection and Immunity 80, 3161-3169.
Fridkin, S.K., Hageman, J.C., Morrison, M., Sanza, L.T., Como-Sabetti, K., Jernigan, J.A.,
Harriman, K., Harrison, L.H., Lynfield, R., and Farley, M.M. (2005). Methicillin-resistant
Staphylococcus aureus disease in three communities. N Engl J Med 352, 1436-1444.
Fuda, C., Suvorov, M., Vakulenko, S.B., and Mobashery, S. (2004). The basis for resistance to
beta-lactam antibiotics by penicillin-binding protein 2a of methicillin-resistant Staphylococcus
aureus. J Biol Chem 279, 40802-40806.
Garofalo, A., Giai, C., Lattar, S., Gardella, N., Mollerach, M., Kahl, B.C., Becker, K., Prince,
A.S., Sordelli, D.O., and Gomez, M.I. (2012). The Length of the Staphylococcus aureus Protein
A Polymorphic Region Regulates Inflammation: Impact on Acute and Chronic Infection. Journal
of Infectious Diseases 206, 81-90.
Geiger, T., Goerke, C., Mainiero, M., Kraus, D., and Wolz, C. (2008). The virulence regulator
sae of Staphylococcus aureus: Promoter activities and response to phagocytosis-related signals. J
Bacteriol 190, 3419-3428.
Genestier, A.L., Michallet, M.C., Prevost, G., Bellot, G., Chalabreysse, L., Peyrol, S., Thivolet,
F., Etienne, J., Lina, G., Vallette, F.M., et al. (2005). Staphylococcus aureus Panton-Valentine
108
leukocidin directly targets mitochondria and induces Bax-independent apoptosis of human
neutrophils. J Clin Invest 115, 3117-3127.
Gillaspy, A.F., Hickmon, S.G., Skinner, R.A., Thomas, J.R., Nelson, C.L., and Smeltzer, M.S.
(1995). Role of the Accessory Gene Regulator (Agr) in Pathogenesis of Staphylococcal
Osteomyelitis. Infect Immun 63, 3373-3380.
Gillet, Y., Issartel, B., Vanhems, P., Fournet, J.C., Lina, G., Bes, M., Vandenesch, F., Piemont,
Y., Brousse, N., Floret, D., et al. (2002). Association between Staphylococcus aureus strains
carrying gene for Panton-Valentine leukocidin and highly lethal necrotising pneumonia in young
immunocompetent patients. Lancet 359, 753-759.
Giraudo, A.T., Mansilla, C., Chan, A., Raspanti, C., and Nagel, R. (2003). Studies on the
expression of regulatory locus sae in Staphylococcus aureus. Curr Microbiol 46, 246-250.
Gomez, M.I., Lee, A., Reddy, B., Muir, A., Soong, G., Pitt, A., Cheung, A., and Prince, A.
(2004). Staphylococcus aureus protein A induces airway epithelial inflammatory responses by
activating TNFR1. Nat Med 10, 842-848.
Gonzalez, B.E., Rueda, A.M., Shelburne, S.A., Musher, D.M., Hamill, R.J., and Hulten, K.G.
(2006). Community-associated strains of methicillin-resistant Staphylococccus aureus as the
cause of healthcare-associated infection. Infect Cont Hosp Ep 27, 1051-1056.
Gouaux, E., Hobaugh, M., and Song, L.Z. (1997). alpha-Hemolysin, gamma-hemolysin, and
leukocidin from Staphylococcus aureus: Distant in sequence but similar in structure. Protein Sci
6, 2631-2635.
109
Graves, S.F., Kobayashi, S.D., Braughton, K.R., Diep, B.A., Chambers, H.F., Otto, M., and
Deleo, F.R. (2010). Relative contribution of Panton-Valentine leukocidin to PMN plasma
membrane permeability and lysis caused by USA300 and USA400 culture supernatants.
Microbes Infect 12, 446-456.
Hadler, J.L., Petit, S., Mandour, M., and Cartter, M.L. (2012). Trends in invasive infection with
methicillin-resistant Staphylococcus aureus, Connecticut, USA, 2001-2010. Emerging infectious
diseases 18, 917-924.
Hamilton, S.M., Bryant, A.E., Carroll, K.C., Lockary, V., Ma, Y., McIndoo, E., Miller, L.G.,
Perdreau-Remington, F., Pullman, J., Risi, G.F., et al. (2007). In vitro production of panton-
valentine leukocidin among strains of methicillin-resistant Staphylococcus aureus causing
diverse infections. Clin Infect Dis 45, 1550-1558.
Haque, N.Z., Zuniga, L.C., Peyrani, P., Reyes, K., Lamerato, L., Moore, C.L., Patel, S., Allen,
M., Peterson, E., Wiemken, T., et al. (2010). Relationship of vancomycin minimum inhibitory
concentration to mortality in patients with methicillin-resistant Staphylococcus aureus hospital-
acquired, ventilator-associated, or health-care-associated pneumonia. Chest 138, 1356-1362.
Herbert, S., Barry, P., and Novick, R.P. (2001). Subinhibitory clindamycin differentially inhibits
transcription of exoprotein genes in Staphylococcus aureus. Infection and Immunity 69, 2996-
3003.
Hidayat, L.K., Hsu, D.I., Quist, R., Shriner, K.A., and Wong-Beringer, A. (2006). High-dose
vancomycin therapy for methicillin-resistant Staphylococcus aureus infections - Efficacy and
toxicity. Archives of Internal Medicine 166, 2138-2144.
110
Hiramatsu, K., Watanabe, S., Takeuchi, F., Ito, T., and Baba, T. (2004). Genetic characterization
of methicillin-resistant Staphylococcus aureus. Vaccine 22, S5-S8.
Ho, P.L., Lo, P.Y., Chow, K.H., Lau, E.H.Y., Lai, E.L., Cheng, V.C.C., and Kao, R.Y. (2010).
Vancomycin MIC creep in MRSA isolates from 1997 to 2008 in a healthcare region in Hong
Kong. J Infection 60, 140-145.
Hongo, I., Baba, T., Oishi, K., Morimoto, Y., Ito, T., and Hiramatsu, K. (2009). Phenol-soluble
modulin alpha 3 enhances the human neutrophil lysis mediated by Panton-Valentine leukocidin.
J Infect Dis 200, 715-723.
Horan, T.C., Andrus, M., and Dudeck, M.A. (2008). CDC/NHSN surveillance definition of
health care-associated infection and criteria for specific types of infections in the acute care
setting. Am J Infect Control 36, 309-332.
Hruz, P., Zinkernagel, A.S., Jenikova, G., Botwin, G.J., Hugot, J.P., Karin, M., Nizet, V., and
Eckmann, L. (2009). NOD2 contributes to cutaneous defense against Staphylococcus aureus
through alpha-toxin-dependent innate immune activation. P Natl Acad Sci USA 106, 12873-
12878.
Huang, H., Flynn, N.M., Kim, J.H., Monchaud, C., Morita, M., and Cohen, S.H. (2006).
Comparisons of community-associated methicillin-resistant Staphylococcus aureus (MRSA) and
hospital-associated MSRA infections in Sacramento, California. J Clin Microbiol 44, 2423-2427.
Hunt, C., Dionne, M., Delorme, M., Murdock, D., Erdrich, A., Wolsey, D., Groom, A., Cheek,
J., Jacobson, J., Cunningham, B., et al. (1999). Four pediatric deaths from community-acquired
111
methicillin-resistant Staphylococcus aureus - Minnesota and North Dakota, 1997-1999. Jama-J
Am Med Assoc 282, 1123-1125.
Ito, T., Hiramatsu, K., Oliveira, D.C., de Lencastre, H., Zhang, K.Y., Westh, H., O'Brien, F.,
Giffard, P.M., Coleman, D., Tenover, F.C., et al. (2009). Classification of Staphylococcal
Cassette Chromosome mec (SCCmec): Guidelines for Reporting Novel SCCmec Elements.
Antimicrob Agents Ch 53, 4961-4967.
Johnson, J.K., Khoie, T., Shurland, S., Kreisel, K., Stine, O.C., and Roghmann, M.C. (2007).
Skin and soft tissue infections caused by methicillin-resistant Staphylococcus aureus USA300
clone. Emerging Infectious Diseases 13, 1195-1200.
Jones, R.N., Farrell, D.J., Mendes, R.E., and Sader, H.S. (2011). Comparative ceftaroline activity
tested against pathogens associated with community-acquired pneumonia: results from an
international surveillance study. The Journal of antimicrobial chemotherapy 66 Suppl 3, iii69-80.
Joo, H.S., Chan, J.L., Cheung, G.Y., and Otto, M. (2010). Subinhibitory concentrations of
protein synthesis-inhibiting antibiotics promote increased expression of the agr virulence
regulator and production of phenol-soluble modulin cytolysins in community-associated
methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 54, 4942-4944.
Kernodle, D.S., Mcgraw, P.A., Barg, N.L., Menzies, B.E., Voladri, R.K.R., and Harshman, S.
(1995). Growth of Staphylococcus-Aureus with Nafcillin in-Vitro Induces Alpha-Toxin
Production and Increases the Lethal Activity of Sterile Broth Filtrates in a Murine Model. J
Infect Dis 172, 410-419.
112
Khatib, R., Ganga, R., Riederer, K., Sharma, M., Fakih, M.G., Johnson, L.B., and Shemes, S.
(2009). Role of SCCmec Type in Outcome of Staphylococcus aureus Bacteremia in a Single
Medical Center. J Clin Microbiol 47, 590-595.
Klevens, R.M., Morrison, M.A., Nadle, J., Petit, S., Gershman, K., Ray, S., Harrison, L.H.,
Lynfield, R., Dumyati, G., Townes, J.M., et al. (2007). Invasive methicillin-resistant
Staphylococcus aureus infections in the United States. Jama 298, 1763-1771.
Koszczol, C., Bernardo, K., Kronke, M., and Krut, O. (2006). Subinhibitory
quinupristin/dalfopristin attenuates virulence of Staphylococcus aureus. J Antimicrob Chemoth
58, 564-574.
Labandeira-Rey, M., Couzon, F., Boisset, S., Brown, E.L., Bes, M., Benito, Y., Barbu, E.M.,
Vazquez, V., Hook, M., Etienne, J., et al. (2007). Staphylococcus aureus Panton-Valentine
leukocidin causes necrotizing pneumonia. Science 315, 1130-1133.
Law, D. (2000). Virulence factors of Escherichia coli O157 and other Shiga toxin-producing E.
coli. J Appl Microbiol 88, 729-745.
Li, M., Cheung, G.Y., Hu, J., Wang, D., Joo, H.S., Deleo, F.R., and Otto, M. (2010).
Comparative analysis of virulence and toxin expression of global community-associated
methicillin-resistant Staphylococcus aureus strains. J Infect Dis 202, 1866-1876.
Li, M., Diep, B.A., Villaruz, A.E., Braughton, K.R., Jiang, X., DeLeo, F.R., Chambers, H.F., Lu,
Y., and Otto, M. (2009). Evolution of virulence in epidemic community-associated methicillin-
resistant Staphylococcus aureus. Proc Natl Acad Sci U S A 106, 5883-5888.
113
Lina, G., Jarraud, S., Ji, G.Y., Greenland, T., Pedraza, A., Etienne, J., Novick, R.P., and
Vandenesch, F. (1998). Transmembrane topology and histidine protein kinase activity of AgrC,
the agr signal receptor in Staphylococcus aureus. Mol Microbiol 28, 655-662.
Liu, G.Y., Essex, A., Buchanan, J.T., Datta, V., Hoffman, H.M., Bastian, J.F., Fierer, J., and
Nizet, V. (2005). Staphylococcus aureus golden pigment impairs neutrophil killing and promotes
virulence through its antioxidant activity. Journal of Experimental Medicine 202, 209-215.
Lodise, T.P., Graves, J., Evans, A., Graffunder, E., Helmecke, M., Lomaestro, B.M., and
Stellrecht, K. (2008). Relationship between vancomycin MIC and failure among patients with
methicillin-resistant Staphylococcus aureus bacteremia treated with vancomycin. Antimicrob
Agents Ch 52, 3315-3320.
Lofblom, J., Kronqvist, N., Uhlen, M., Stahl, S., and Wernerus, H. (2007). Optimization of
electroporation-mediated transformation: Staphylococcus carnosus as model organism. Journal
of Applied Microbiology 102, 736-747.
Malachowa, N., Kobayashi, S.D., and DeLeo, F.R. (2012). Community-associated methicillin-
resistant Staphylococcus aureus and athletes. Phys Sportsmed 40, 13-21.
Maree, C.L., Daum, R.S., Boyle-Vavra, S., Matayoshi, K., and Miller, L.G. (2007). Community-
associated methicillin-resistant Staphylococcus aureus isolates causing healthcare-associated
infections. Emerging Infectious Diseases 13, 236-242.
Martin, F.J., Gomez, M.I., Wetzel, D.M., Memmi, G., O'Seaghdha, M., Soong, G., Schindler, C.,
and Prince, A. (2009). Staphylococcus aureus activates type I IFN signaling in mice and humans
through the Xr repeated sequences of protein A. J Clin Invest 119, 1931-1939.
114
Matamouros, S., England, P., and Dupuy, B. (2007). Clostridium difficile toxin expression is
inhibited by the novel regulator TcdC. Mol Microbiol 64, 1274-1288.
McElroy, M.C., Harty, H.R., Hosford, G.E., Boylan, G.M., Pittet, J.F., and Foster, T.J. (1999).
Alpha-toxin damages the air-blood barrier of the lung in a rat model of Staphylococcus aureus-
induced pneumonia. Infect Immun 67, 5541-5544.
McLoughlin, R.M., Solinga, R.M., Rich, J., Zaleski, K.J., Cocchiaro, J.L., Risley, A., Tzianabos,
A.O., and Lee, J.C. (2006). CD4(+) T cells and CXC chemokines modulate the pathogenesis of
Staphylococcus aureus wound infections. P Natl Acad Sci USA 103, 10408-10413.
Miller, L.G., Perdreau-Remington, F., Bayer, A.S., Diep, B., Tan, N., Bharadwa, K., Tsui, J.,
Perlroth, J., Shay, A., Tagudar, G., et al. (2007). Clinical and epidemiologic characteristics
cannot distinguish community-associated methicillin-resistant Staphylococcus aureus infection
from methicillin-susceptible S. aureus infection: a prospective investigation. Clinical infectious
diseases : an official publication of the Infectious Diseases Society of America 44, 471-482.
Moise, P.A., North, D., Steenbergen, J.N., and Sakoulas, G. (2009). Susceptibility relationship
between vancomycin and daptomycin in Staphylococcus aureus: facts and assumptions. Lancet
Infectious Diseases 9, 617-624.
Mongkolrattanothai, K., Boyle, S., Kahana, M.D., and Daum, R.S. (2003). Severe
Staphylococcus aureus infections caused by clonally related community-acquired methicillin-
susceptible and methicillin-resistant isolates. Clin Infect Dis 37, 1050-1058.
Montgomery, C.P., Boyle-Vavra, S., Adem, P.V., Lee, J.C., Husain, A.N., Clasen, J., and Daum,
R.S. (2008). Comparison of virulence in community-associated methicillin-resistant
115
Staphylococcus aureus pulsotypes USA300 and USA400 in a rat model of pneumonia. J Infect
Dis 198, 561-570.
Moore, C.L., Osaki-Kiyan, P., Haque, N.Z., Perri, M.B., Donabedian, S., and Zervos, M.J.
(2012). Daptomycin Versus Vancomycin for Bloodstream Infections Due to Methicillin-
Resistant Staphylococcus aureus With a High Vancomycin Minimum Inhibitory Concentration:
A Case-Control Study. Clin Infect Dis 54, 51-58.
Moran, G.J., Krishnadasan, A., Gorwitz, R.J., Fosheim, G.E., McDougal, L.K., Carey, R.B., and
Talan, D.A. (2006). Methicillin-resistant S-aureus infections among patients in the emergency
department. New Engl J Med 355, 666-674.
Musta, A.C., Riederer, K., Shemes, S., Chase, P., Jose, J., Johnson, L.B., and Khatib, R. (2009).
Vancomycin MIC plus Heteroresistance and Outcome of Methicillin-Resistant Staphylococcus
aureus Bacteremia: Trends over 11 Years. J Clin Microbiol 47, 1640-1644.
Naimi, T.S., LeDell, K.H., Como-Sabetti, K., Borchardt, S.M., Boxrud, D.J., Etienne, J.,
Johnson, S.K., Vandenesch, F., Fridkin, S., O'Boyle, C., et al. (2003). Comparison of
community- and health care-associated methicillin-resistant Staphylococcus aureus infection.
Jama-J Am Med Assoc 290, 2976-2984.
Nilsson, I.M., Bremell, T., Ryden, C., Cheung, A.L., and Tarkowski, A. (1996). Role of the
staphylococcal accessory gene regulator (sar) in septic arthritis. Infect Immun 64, 4438-4443.
Novick, R.P. (2003). Autoinduction and signal transduction in the regulation of staphylococcal
virulence. Mol Microbiol 48, 1429-1449.
116
Novick, R.P., and Jiang, D.R. (2003). The staphylococcal saeRS system coordinates
environmental signals with agr quorum sensing. Microbiol-Sgm 149, 2709-2717.
Novick, R.P., Ross, H.F., Projan, S.J., Kornblum, J., Kreiswirth, B., and Moghazeh, S. (1993).
Synthesis of Staphylococcal Virulence Factors Is Controlled by a Regulatory Rna Molecule.
Embo J 12, 3967-3975.
Nygaard, T.K., Pallister, K.B., Ruzevich, P., Griffith, S., Vuong, C., and Voyich, J.M. (2010).
SaeR Binds a Consensus Sequence within Virulence Gene Promoters to Advance USA300
Pathogenesis. Journal of Infectious Diseases 201, 241-254.
Ohlsen, K., Ziebuhr, W., Koller, K.P., Hell, W., Wichelhaus, T.A., and Hacker, J. (1998). Effects
of subinhibitory concentrations of antibiotics on alpha-toxin (hla) gene expression of methicillin-
sensitive and methicillin-resistant Staphylococcus aureus isolates. Antimicrob Agents Chemother
42, 2817-2823.
Olson, M.W., Ruzin, A., Feyfant, E., Rush, T.S., O'Connell, J., and Bradford, P.A. (2006).
Functional, biophysical, and structural bases for antibacterial activity of tigecycline. Antimicrob
Agents Ch 50, 2156-2166.
Periasamy, S., Joo, H.S., Duong, A.C., Bach, T.H.L., Tan, V.Y., Chatterjee, S.S., Cheung,
G.Y.C., and Otto, M. (2012). How Staphylococcus aureus biofilms develop their characteristic
structure. P Natl Acad Sci USA 109, 1281-1286.
Peschel, A., Jack, R.W., Otto, M., Collins, L.V., Staubitz, P., Nicholson, G., Kalbacher, H.,
Nieuwenhuizen, W.F., Jung, G., Tarkowski, A., et al. (2001). Staphylococcus aureus resistance
to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is
117
based on modification of membrane lipids with L-lysine. Journal of Experimental Medicine 193,
1067-1076.
Peschel, A., Otto, M., Jack, R.W., Kalbacher, H., Jung, G., and Gotz, F. (1999). Inactivation of
the dlt operon in Staphylococcus aureus confers sensitivity to defensins, protegrins, and other
antimicrobial peptides. J Biol Chem 274, 8405-8410.
Peyrani, P., Allen, M., Wiemken, T.L., Haque, N.Z., Zervos, M.J., Ford, K.D., Scerpella, E.G.,
Mangino, J.E., Kett, D.H., and Ramirez, J.A. (2011). Severity of disease and clinical outcomes in
patients with hospital-acquired pneumonia due to methicillin-resistant Staphylococcus aureus
strains not influenced by the presence of the Panton-Valentine leukocidin gene. Clinical
infectious diseases : an official publication of the Infectious Diseases Society of America 53,
766-771.
Pinho, M.G., De Lencastre, H., and Tomasz, A. (2001). An acquired and a native penicillin-
binding protein cooperate in building the cell wall of drug-resistant and drug-susceptible
staphylococci. Clin Infect Dis 33, 1090-1090.
Prince, L.R., Graham, K.J., Connolly, J., Anwar, S., Ridley, R., Sabroe, I., Foster, S.J., and
Whyte, M.K.B. (2012). Staphylococcus aureus Induces Eosinophil Cell Death Mediated by
alpha-hemolysin. Plos One 7.
Queck, S.Y., Jameson-Lee, M., Villaruz, A.E., Bach, T.H.L., Khan, B.A., Sturdevant, D.E.,
Ricklefs, S.M., Li, M., and Otto, M. (2008). RNAIII-Independent Target Gene Control by the agr
Quorum-Sensing System: Insight into the Evolution of Virulence Regulation in Staphylococcus
aureus. Mol Cell 32, 150-158.
118
Ragle, B.E., and Bubeck Wardenburg, J. (2009). Anti-alpha-hemolysin monoclonal antibodies
mediate protection against Staphylococcus aureus pneumonia. Infection and Immunity 77, 2712-
2718.
Rautenberg, M., Joo, H.S., Otto, M., and Peschel, A. (2011). Neutrophil responses to
staphylococcal pathogens and commensals via the formyl peptide receptor 2 relates to phenol-
soluble modulin release and virulence. Faseb J 25, 1254-1263.
Rechtin, T.M., Gillaspy, A.F., Schumacher, M.A., Brennan, R.G., Smeltzer, M.S., and Hurlburt,
B.K. (1999). Characterization of the SarA virulence gene regulator of Staphylococcus aureus.
Mol Microbiol 33, 307-316.
Rigby, K.M., and DeLeo, F.R. (2012). Neutrophils in innate host defense against Staphylococcus
aureus infections. Seminars in immunopathology 34, 237-259.
Rubinstein, E., Kollef, M.H., and Nathwani, D. (2008). Pneumonia caused by methicillin-
resistant Staphylococcus aureus. Clin Infect Dis 46, S378-S385.
Rybak, M., Lomaestro, B., Rotschafer, J.C., Moellering, R., Jr., Craig, W., Billeter, M.,
Dalovisio, J.R., and Levine, D.P. (2009). Therapeutic monitoring of vancomycin in adult
patients: a consensus review of the American Society of Health-System Pharmacists, the
Infectious Diseases Society of America, and the Society of Infectious Diseases Pharmacists. Am
J Health Syst Pharm 66, 82-98.
Sakoulas, G., Moise-Broder, P.A., Schentag, J., Forrest, A., Moellering, R.C., and Eliopoulos,
G.M. (2004). Relationship of MIC and bactericidal activity to efficacy of vancomycin for
119
treatment of methicillin-resistant Staphylococcus aureus bacteremia. J Clin Microbiol 42, 2398-
2402.
Schlievert, P.M., and Kelly, J.A. (1984). Clindamycin-Induced Suppression of Toxic-Shock
Syndrome Associated Exotoxin Production. J Infect Dis 149, 471-471.
Schumacher, M.A., Hurlburt, B.K., and Brennan, R.G. (2001). Crystal structures of SarA, a
pleiotropic regulator of virulence genes in S-aureus. Nature 409, 215-219.
Schwan, W.R., Langhorne, M.H., Ritchie, H.D., and Stover, C.K. (2003). Loss of hemolysin
expression in Staphylococcus aureus agr mutants correlates with selective survival during mixed
infections in murine abscesses and wounds. FEMS Immunol Med Microbiol 38, 23-28.
Seybold, U., Halvosa, J.S., White, N., Voris, V., Ray, S.M., and Blumberg, H.M. (2008).
Emergence of and Risk Factors for Methicillin-Resistant Staphylococcus aureus of Community
Origin in Intensive Care Nurseries. Pediatrics 122, 1039-1046.
Seybold, U., Kourbatova, E.V., Johnson, J.G., Halvosa, S.J., Wang, Y.F., King, M.D., Ray, S.M.,
and Blumberg, H.M. (2006). Emergence of community-associated methicillin-resistant
Staphylococcus aureus USA300 genotype as a major cause of health care-associated blood
stream infections. Clin Infect Dis 42, 647-656.
Sharma-Kuinkel, B.K., Ahn, S.H., Rude, T.H., Zhang, Y., Tong, S.Y., Ruffin, F., Genter, F.C.,
Braughton, K.R., Deleo, F.R., Barriere, S.L., et al. (2012). Presence of genes encoding panton-
valentine leukocidin is not the primary determinant of outcome in patients with hospital-acquired
pneumonia due to Staphylococcus aureus. J Clin Microbiol 50, 848-856.
120
Soong, G., Martin, F.J., Chun, J., Cohen, T.S., Ahn, D.S., and Prince, A. (2011a).
Staphylococcus aureus protein A mediates invasion across airway epithelial cells through
activation of RhoA GTPase signaling and proteolytic activity. J Biol Chem 286, 35891-35898.
Soong, G., Martin, F.J., Chun, J.R., Cohen, T.S., Ahn, D.S., and Prince, A. (2011b).
Staphylococcus aureus Protein A Mediates Invasion across Airway Epithelial Cells through
Activation of RhoA GTPase Signaling and Proteolytic Activity. J Biol Chem 286, 35891-35898.
Soriano, A., Marco, F., Martinez, J.A., Pisos, E., Almela, M., Dimova, V.P., Alamo, D., Ortega,
M., Lopez, J., and Mensa, J. (2008). Influence of vancomycin minimum inhibitory concentration
on the treatment of methicillin-resistant Staphylococcus aureus bacteremia. Clin Infect Dis 46,
193-200.
Steinhuber, A., Goerke, C., Bayer, M.G., Doring, G., and Wolz, C. (2003). Molecular
architecture of the regulatory locus sae of Staphylococcus aureus and its impact on expression of
virulence factors. J Bacteriol 185, 6278-6286.
Stevens, D.L., Ma, Y., Salmi, D.B., McIndoo, E., Wallace, R.J., and Bryant, A.E. (2007a).
Impact of antibiotics on expression of virulence-associated exotoxin genes in methicillin-
sensitive and methicillin-resistant Staphylococcus aureus. Journal of Infectious Diseases 195,
202-211.
Stevens, D.L., Ma, Y.S., Salmi, D.B., McIndoo, E., Wallace, R.J., and Bryant, A.E. (2007b).
Impact of antibiotics on expression of virulence-associated exotoxin genes in methicillin-
sensitive and methicillin-resistant Staphylococcus aureus. J Infect Dis 195, 202-211.
121
Styers, D., Sheehan, D.J., Hogan, P., and Sahm, D.F. (2006). Laboratory-based surveillance of
current antimicrobial resistance patterns and trends among Staphylococcus aureus: 2005 status in
the United States. Ann Clin Microbiol Antimicrob 5, 2.
Sun, F., Li, C.L., Jeong, D., Sohn, C., He, C., and Bae, T. (2010). In the Staphylococcus aureus
Two-Component System sae, the Response Regulator SaeR Binds to a Direct Repeat Sequence
and DNA Binding Requires Phosphorylation by the Sensor Kinase SaeS. J Bacteriol 192, 2111-
2127.
Swaney, S.M., Aoki, H., Ganoza, M.C., and Shinabarger, D.L. (1998). The oxazolidinone
linezolid inhibits initiation of protein synthesis in bacteria. Antimicrob Agents Ch 42, 3251-
3255.
Tacconelli, E. (2009). Methicillin-resistant Staphylococcus aureus: source control and
surveillance organization. Clin Microbiol Infec 15, 31-38.
Tenover, F.C., McDougal, L.K., Goering, R.V., Killgore, G., Projan, S.J., Patel, J.B., and
Dunman, P.M. (2006). Characterization of a strain of community-associated methicillin-resistant
Staphylococcus aureus widely disseminated in the United States. J Clin Microbiol 44, 108-118.
Tenson, T., Lovmar, M., and Ehrenberg, M. (2003). The mechanism of action of macrolides,
lincosamides and streptogramin B reveals the nascent peptide exit path in the ribosome. J Mol
Biol 330, 1005-1014.
Trotonda, M.P., Xiong, Y.Q., Memmi, G., Bayer, A.S., and Cheung, A.L. (2009). Role of mgrA
and sarA in Methicillin-Resistant Staphylococcus aureus Autolysis and Resistance to Cell Wall-
Active Antibiotics. J Infect Dis 199, 209-218.
122
van Hal, S.J., and Paterson, D.L. (2011). Systematic review and meta-analysis of the significance
of heterogeneous vancomycin-intermediate Staphylococcus aureus isolates. Antimicrob Agents
Ch 55, 405-410.
Vandenesch, F., Naimi, T., Enright, M.C., Lina, G., Nimmo, G.R., Heffernan, H., Liassine, N.,
Bes, M., Greenland, T., Reverdy, M.E., et al. (2003). Community-acquired methicillin-resistant
Staphylococcus aureus carrying Panton-Valentine leukocidin genes: Worldwide emergence.
Emerging Infectious Diseases 9, 978-984.
vanLangevelde, P., vanDissel, J.T., Meurs, C.J.C., Renz, J., and Groeneveld, P.H.P. (1997).
Combination of flucloxacillin and gentamicin inhibits toxic shock syndrome toxin 1 production
by Staphylococcus aureus in both logarithmic and stationary phases of growth. Antimicrob
Agents Ch 41, 1682-1685.
Wang, G., Clark, C.G., and Rodgers, F.G. (2002). Detection in Escherichia coli of the genes
encoding the major virulence factors, the genes defining the O157:H7 serotype, and components
of the type 2 Shiga toxin family by multiplex PCR. J Clin Microbiol 40, 3613-3619.
Wang, G.Q., Hindler, J.F., Ward, K.W., and Bruckner, D.A. (2006). Increased vancomycin MICs
for Staphylococcus aureus clinical isolates from a university hospital during a 5-year period. J
Clin Microbiol 44, 3883-3886.
Wang, R., Braughton, K.R., Kretschmer, D., Bach, T.H.L., Queck, S.Y., Li, M., Kennedy, A.D.,
Dorward, D.W., Klebanoff, S.J., Peschel, A., et al. (2007). Identification of novel cytolytic
peptides as key virulence determinants for community-associated MRSA. Nat Med 13, 1510-
1514.
123
Wardenburg, J.B., Bae, T., Otto, M., Deleo, F.R., and Schneewind, O. (2007). Poring over pores:
alpha-hemolysin and Panton-Valentine leukocidin in Staphylococcus aureus pneumonia. Nat
Med 13, 1405-1406.
Wardenburg, J.B., and Schneewind, O. (2008). Vaccine protection against Staphylococcus
aureus pneumonia. J Exp Med 205, 287-294.
Werber, D., Fruth, A., Buchholz, U., Prager, R., Kramer, M.H., Ammon, A., and Tschape, H.
(2003). Strong association between shiga toxin-producing Escherichia coli O157 and virulence
genes stx2 and eae as possible explanation for predominance of serogroup O157 in patients with
haemolytic uraemic syndrome. Eur J Clin Microbiol Infect Dis 22, 726-730.
Widaa, A., Claro, T., Foster, T.J., O'Brien, F.J., and Kerrigan, S.W. (2012). Staphylococcus
aureus Protein A Plays a Critical Role in Mediating Bone Destruction and Bone Loss in
Osteomyelitis. Plos One 7.
Xiang, M., and Fan, J. (2010). Pattern recognition receptor-dependent mechanisms of acute lung
injury. Mol Med 16, 69-82.
Xiong, Y.Q., Willard, J., Yeaman, M.R., Cheung, A.L., and Bayer, A.S. (2006). Regulation of
Staphylococcus aureus alpha-toxin gene (hla) expression by agr, sarA, and sae in vitro and in
experimental infective endocarditis. Journal of Infectious Diseases 194, 1267-1275.
Yamaki, J., Lee, M., Shriner, K.A., and Wong-Beringer, A. (2011a). Can clinical and molecular
epidemiologic parameters guide empiric treatment with vancomycin for methicillin-resistant
Staphylococcus aureus infections? Diagn Micr Infec Dis 70, 124-130.
124
Yamaki, J., Synold, T., and Wong-Beringer, A. (2011b). Antivirulence Potential of TR-700 and
Clindamycin on Clinical Isolates of Staphylococcus aureus Producing Phenol-Soluble Modulins.
Antimicrob Agents Ch 55, 4432-4435.
Yanagihara, K., Morinaga, Y., Nakamura, S., Seki, M., Izumikawa, K., Kakeya, H., Yamamoto,
Y., Yamada, Y., Kamihira, S., and Kohno, S. (2008). Subinhibitory concentrations of
telithromycin, clarithromycin and azithromycin reduce methicillin-resistant Staphylococcus
aureus coagulase in vitro and in vivo. J Antimicrob Chemoth 61, 647-650.
Yarovinsky, T.O. (2012). Role of phospholipid scramblase 1 in type I interferon-induced
protection from staphylococcal alpha-toxin. Virulence 3, 457-458.
Zervos, M.J., Chua, T., Moore, C.L., Perri, M.B., Donabedian, S.M., Masch, W., Vager, D.,
Davis, S.L., Lulek, K., and Zimnicki, B. (2008). Molecular epidemiology of methicillin-resistant
Staphylococcus aureus bloodstream isolates in urban Detroit. J Clin Microbiol 46, 2345-2352.
Zetola, N., Francis, J.S., Nuermberger, E.L., and Bishai, W.R. (2005). Community-acquired
meticillin-resistant Staphylococcus aureus: an emerging threat. Lancet Infect Dis 5, 275-286.
Abstract (if available)
Abstract
Methicillin-resistant Staphylococcus aureus (MRSA) has become a major public health problem due to the high prevalence of infections caused by these strains in the U.S. and worldwide. MRSA is now the leading cause of skin soft tissue infections presented to emergency departments (Moran et al., 2006) and a cause of 94,000 estimated cases of invasive infections per year in the U.S. (Klevens et al., 2007). Adding to this challenge is the emergence of bacterial strains resistant to standard treatment and unique virulent clones that affect otherwise healthy individuals in the community. Molecular epidemiologic studies have described increasing involvement of MRSA strains originated from the community (CA) as the cause of healthcare-associated (HA) infections in other parts of the country. ❧ The main focus of the studies described in this thesis was to explore the epidemiology of MRSA strains within our institution and identify bacterial or patient markers that can be used to guide treatment and improve outcomes. Furthermore, we investigated whether observations described in other bacterial toxin-mediated disease is applicable to MRSA infection, including correlation of in-vitro exotoxin production to disease severity and treatment approaches to inhibit toxin production. ❧ We found significant changes in the molecular epidemiology of MRSA within our institution. Based on this change we offer possible markers that can be used to guide empiric vancomycin treatment. We also present a timeframe by which patient response should be evaluated and the possibility of switching from vancomycin to an alternative agent. Furthermore, in our studies we found the in-vitro toxin production and decreased vancomycin susceptibility did not have major impacts on disease severity and poor outcomes. Finally, we found that under sub-inhibitory concentrations of protein synthesis inhibiting antibiotics PSM peptides can be induced contrary to previous studies examining other exotoxins. This data challenges the assumption that protein synthesis inhibiting antibiotics used in the treatment of MRSA infections will only add additional benefits with no risks.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Antibiotic immunomodulation in Staphylococcus aureus bloodstream infection
PDF
Biological impact of fluoroquinolone resistance in Pseudomonas aeruginosa
PDF
Therapeutic potential of rhesus theta defensin-1 in the treatment of inflammatory lung diseases
PDF
Understanding the contribution of microbial resistance and virulence towards outcomes of Staphylococcus aureus bacteremia
PDF
Therapeutic potential of host defense peptides for the treatment of cystic fibrosis lung disease
PDF
Investigation of in vitro activity and in vivo safety of imipenem-relebactam alone and in combination with other anti-microbial agents against clinical isolates of M. abscessus from CF patients
PDF
Pharmacokinetic and pharmacodynamic optimization of CFTR modulator therapy to mitigate potential drug interactions and adverse events in people with cystic fibrosis
PDF
Neuroinflammation and ApoE4 genotype in at-risk female aging: implications for Alzheimer's disease
PDF
Assessing the distribution and role of the CRISPR-Cas system on carbapenem resistance in e. coli and k. pneumoniae clinical strains
PDF
Renin-angiotensin system modulation for the prevention and treatment of metabolic dysfunction
PDF
Averting dementia: renin-angiotensin system and angiogenic cells in cognitive decline
PDF
Therapeutic potential of rhesus theta (θ) defensin-1 in cystic fibrosis airway infection and inflammation
PDF
Developing ultrasensitive ethanol receptors (USERS) as novel tools for alcohol research: optimizing Loop 2 mutations in α1 GlyRs
PDF
Preclinical investigation of ivermectin as a novel therapeutic agent for treatment of alcohol use disorders
PDF
Utilization of a THP-1 macrophage model to test efficacy of single drug and combination therapy against mycobacterium abscessus
PDF
Screening of cysteine-rich backbone cyclized peptide libraries for inhibitors against disease-related proteases
PDF
Wastewater-based epidemiology for emerging biological contaminants
PDF
Development of dihydromyricetin (DHM) as a novel therapy for alcoholic liver disease (ALD) and alcohol use disorder (AUD)
PDF
From risk mitochondrial and metabolic phenotype towards a precision medicine approach for Alzheimer's disease
PDF
Beneficial effect of antibiotic treatment on alcohol-related liver pathology in mice is not due to reduction in the butyrate-producing gut microbial phyla
Asset Metadata
Creator
Yamaki, Jason
(author)
Core Title
Methicillin-resistant Staphylococcus aureus: molecular epidemiology, virulence in disease, and antibiotic modulation of virulence
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Clinical and Experimental Therapeutics
Publication Date
08/05/2013
Defense Date
05/15/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
antibiotics,MRSA,OAI-PMH Harvest,Staphylococcus aureus,virulence
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Alkana, Ronald L. (
committee chair
), Rodgers, Kathleen E. (
committee chair
), Wong-Beringer, Annie (
committee chair
), Ouellette, Andre J. (
committee member
)
Creator Email
jayou812@gmail.com,jmyamaki@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-314652
Unique identifier
UC11293844
Identifier
etd-YamakiJaso-1948.pdf (filename),usctheses-c3-314652 (legacy record id)
Legacy Identifier
etd-YamakiJaso-1948.pdf
Dmrecord
314652
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Yamaki, Jason
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
antibiotics
MRSA
virulence