1. Introduction
Ventilator-associated pneumonia (VAP) is a type of nosocomial pneumonia that affects 9 to 27% of patients subjected to mechanical ventilation [
1,
2]. Mortality attributed to VAP varies according to the risk group studied, with surgical and long-stay patients being more likely to die from the disease [
3]. Thus, the epidemiological and molecular study of factors related to the development of VAP is of great importance since it can help identify risk groups and develop new interventions.
The most common etiological agents of VAP are aerobic Gram-positive bacteria such as
Staphylococcus aureus and Gram-negative bacteria such as
Pseudomonas aeruginosa,
Acinetobacter baumannii, and
Klebsiella pneumoniae.
S. aureus is the main causative agent of nosocomial pneumonia and the second most common agent associated with all nosocomial infections [
4,
5].
Pneumonias caused by methicillin-resistant
S. aureus (MRSA) are generally difficult to treat and are associated with high rates of treatment failure [
6]. MRSA strains carry the
mecA gene, which confers resistance to beta-lactam antibiotics and is located on a mobile genetic element (staphylococcal cassette chromosome
mec—SCC
mec) of variable size and genetic composition. SCC
mec typing is widely applied in epidemiological studies and can be useful for clinical purposes. There are currently 14 SCC
mec types (I to XIV) [
7]. Regarding their epidemiological distribution, types I to III are mostly found in hospitals, and the other types are predominantly community-associated. Haque et al. [
6] found that SCC
mec type II in conjunction with host factors is associated with mortality, high vancomycin minimum inhibitory concentrations (MIC), and clinical treatment failure in patients with nosocomial pneumonia caused by MRSA. This finding highlights the importance of epidemiological–molecular research on these microorganisms associated with VAP.
Staphylococcus aureus has a great ability to invade host tissues due to the production of different virulence factors. Several studies have addressed the role that each virulence factor plays in the severity of pneumonia and/or lung injury, which range from the investigation of bacterial cytotoxicity [
6,
8] to exploring the role of each virulence gene. Among the virulence factors that play an important role in pneumonia, alpha-hemolysin or alpha-toxin (Hla) is one of the most prominent cytotoxins that target erythrocytes, epithelial and endothelial cells, T cells, neutrophils, monocytes, and macrophages. Alpha-toxin is an amphipathic, water-soluble, pore-forming molecule that binds primarily to lipids present on the host cell membrane [
9,
10]. The main function of alpha-toxin is the causing osmotic swelling, rupture, lysis, and subsequently, cell death. It is highly inflammatory, causing pulmonary congestion [
11].
Other factors, such as staphylococcal enterotoxins (SEs) and toxic shock syndrome toxin 1 (TSST-1), are superantigens involved in diverse human infections. To explore their participation in pneumonia, experimental studies using rabbits have shown that intrapulmonary instillation of SEB and SEC induces hemorrhage in the pulmonary tissue, causing symptoms of respiratory distress and lethal toxic shock syndrome. The administration of strains carrying the genes for the same enterotoxins resulted in pulmonary pathology and lethality similar to the group that received the purified toxins [
12,
13].
The participation of Panton–Valentine leukocidin (PVL) in pneumonia has been extensively reported in the literature, especially in pneumonia caused by community-acquired
S. aureus strains [
14,
15]. PVL is a pore-forming toxin that targets neutrophils and mitochondria. It induces cell lysis, thereby releasing proinflammatory agents into the extracellular environment that cause tissue necrosis. When present in the lung, PVL is associated with necrotizing pneumonia [
16]. Some strains that carry the PVL gene also harbor genes encoding exfoliative toxins (
eta,
etb, or
etd) [
17]. The role of exfoliative toxins in skin infections is well understood, but little is known about their participation in pneumonia; hence, research on
S. aureus-associated pneumonia is needed.
Another important virulence factor in the pathogenesis of
S. aureus pneumonia is biofilm formation, particularly in the area of the endotracheal tube. The
icaADBC-mediated polysaccharide production is an important mechanism for biofilm formation and contributes to the early growth of bacteria [
18]. The
bap gene of
S. aureus encodes a surface protein Bap (biofilm-associated protein) containing 2276 amino acids. Bap was identified as the main determinant of successful surface adhesion and intercellular adhesion during biofilm formation [
18].
In addition to bacterial factors, the involvement of host-related factors in the development of VAP has been widely studied. The use of antimicrobials and antacids, opting for tracheostomy, supine position, parenteral nutrition, AIDS, male gender, lung disease, coma, and trauma are risk factors for the development of VAP [
19].
In view of the above considerations, our study aimed to explore the molecular epidemiology of S. aureus isolated from patients on mechanical ventilation and to investigate the participation of virulence and patient-related factors in the development of VAP.
3. Discussion
Nosocomial pneumonias, together with surgical site infections, are the main types of nosocomial infection, and
S. aureus is the leading associated etiological agent, according to a survey conducted in the United States [
4]. The present study showed an ID of
S. aureus VAP of 2.35/1000 ventilator days, a value higher than that reported in the study by Lee et al. [
20] in which the ID of VAP was 1.4/1000 ventilator days and
S. aureus was the most frequently isolated microorganism.
In our study, the presence of the
mecA gene, oxacillin resistance, or SCC
mec type was not associated with the development of VAP or death. Some studies have shown an association between infection with MRSA methicillin-resistant strains and a longer ICU stay and higher mortality rates compared to patients who develop VAP caused by MSSA [
21,
22]. However, other studies found that methicillin resistance alone was not associated with recurrence, severity, or mortality in
S. aureus VAP [
23,
24]. Thus, methicillin resistance does not seem to interfere with the mortality rate in VAP, a fact that raises speculation about whether other factors present in nosocomial strains may contribute to mortality. Interestingly, the frequency of the
icaC,
sec,
sea, and
hlb genes was slightly higher in
mecA gene-positive isolates; however, there was no statistically significant association between the simultaneous presence of the
mecA gene and virulence genes, with MSSA and MRSA isolates showing the same level of virulence.
The wide variability in the clonal profiles among strains isolated from patients with VAP and the formation of clusters only for isolates from the same individual suggest that these isolates are endogenous and that their development occurs from strains that already colonize the patient. Only two isolates (47 and 239) that belonged to the same clone formed a cluster. Previous studies have shown that the pattern of microorganism colonization varies. Gram-positive bacteria colonize the trachea within the first 24 h of mechanical ventilation and are later replaced with Gram-negative bacteria and yeasts [
25,
26].
Sequence types characteristic of community settings were the types most frequently isolated from patients who developed VAP (ST45, ST5, ST8, ST1635, ST398, and SLV 546), with the exception of nosocomial clone ST105 carrying SCC
mec II. One strain was grouped with USA300, which belonged to the same ST8 carrying SCC
mec IV, and was negative for the PVL (
lukSF-PV) genes. Classic studies of infection with USA300 mainly report necrotizing pneumonia [
27] in community-dwelling healthy individuals without traditional factors for MRSA contraction [
28]. Despite reports of pneumonia in community-dwelling patients, Pasquale et al. [
29] recently described the involvement of USA300 in nosocomial pneumonia. In these cases, participation of this strain is of great clinical importance since, in addition to the difficulty in selecting antimicrobials for treatment, the strain can carry a wide range of virulence factors that can also influence the outcome [
3,
9].
The involvement of STs characteristic of community settings in nosocomial infections has been reported, especially an association of ST8 with outbreaks of difficult-to-treat infections [
30]. This observation highlights the need for controlling the transmission of these community-associated strains among patients admitted to the same ICU.
Studies have discussed the participation of virulence factors in the development of VAP in an attempt to identify a marker that permits predicting the development and severity of these infections. The participation of a single virulence factor such as alpha-hemolysin or protein A has been reported to be important, if not crucial, for the development of pneumonia [
9,
31].
A previous study also found that patients with cancer, especially lung cancer, are at increased risk of developing VAP when compared to other patients [
32]. The presence of
sea in the lungs can elicit an inflammatory response through the production of IFN-γ by CD8 T cells, which causes lesions characteristic of acute respiratory stress syndrome [
33]. This finding suggests that strains producing this enterotoxin pose a risk for developing pneumonia associated with severe symptoms.
The only independent factors associated with death were the presence of the
sea and
hlb genes. Among the four
S. aureus strains isolated from patients with VAP who died from this condition, three carried the
hlb gene and the other the
sea gene. The presence of enterotoxin A has been linked to aggressive infection [
33]. Similarly, the presence of
hlb in lung tissue can increase neutrophil infiltration by modulating host factors, which contributes to the development of pneumonia [
34]. Both death-related factors are known to activate an intense immune response that contributes to breathing problems and can quickly lead to the patient’s death.
The association between the presence of virulence genes and the development of VAP has been little explored, and studies are limited to the analysis of factors such as strain cytotoxicity [
8] or few virulence factors in animal experiments [
6,
31], which offers an excellent perspective of the virulence factors studied. However, our study provides a practical view of virulence factors and new perspectives on the association of microbiological factors with VAP. Within this context, we found that the etiology of VAP is multifactorial, including microbiological factors and risk factors of the host itself.
The main strength of this study is the fact that the patients were followed up since intubation, which allows microbiological study and the analysis of clinical data throughout the period of development of VAP. The results suggest that, in patients admitted to the ICU, VAP can be caused by S. aureus strains that already colonize these patients. This suggestion is supported by the STs identified, which are characteristic of strains found in the community that carry important virulence factors involved in the development of VAP and subsequent death of the patient. The identification of patients with VAP previously colonized with S. aureus suggests that some type of screening may be useful in predicting the occurrence of infection. However, further studies are still needed to evaluate the best screening approach and its cost-effectiveness. In view of the high mortality associated with VAP, the results of this study also highlight the need for future studies that assess whether S. aureus decolonization of known carriers will affect the incidence of VAP.
4. Materials and Methods
4.1. Study Place and Sample
The study was conducted in two Specialized Intensive Care Units (ICU) of the University Hospital of the Botucatu Medical School (HC-FMB), with 16 and 9 beds, respectively, which admit adult clinical and surgical patients. Patients meeting the following criteria were included: patients undergoing mechanical ventilation started within 48 h of admission to the ICU or during their stay without a previous diagnosis of pneumonia whose families agreed to their participation in the study by signing the free informed consent form.
This is a prospective cohort study whose primary outcomes are VAP caused by
S. aureus and death due to VAP. The subjects were followed up for a maximum period of eight weeks of ICU stay, twice weekly (Mondays and Thursdays), at an interval of 3.5 days. Clinical data and tracheal aspirates were collected during the evaluations. A doctor of the Committee for the Control of Healthcare-Related Infection (CCIRAS) made the diagnosis of VAP. The criteria recommended by the National Healthcare Safety Network (NHSN) of the Centers for Disease Control and Prevention (CDC) of the United States were used [
35].
4.2. Collection of Clinical Data
The following data were obtained during the visit for patient inclusion: gender, age, underlying diseases and comorbidities, length of hospital stay prior to the current hospitalization, procedures, and presence of invasive devices (central venous catheters, indwelling urinary catheters, nasogastric tubes/nasoenteral tubes, and drains). Data comprising the two weeks preceding mechanical ventilation were also collected: past and current use of antimicrobials, history of infectious complications, and results of microbiological tests.
Data on length of stay, procedures, invasive devices, use of antimicrobials, infectious complications, and results of microbiological tests were also collected during the follow-up assessments.
4.3. Collection, Isolation, and Microbial Identification
Tracheal aspirate was collected by a physical therapist, doctor, or nurse and immediately sent to the Laboratory of Bacteriology. The clinical material was diluted 1:1 in 1% N-acetyl-L-cysteine (mucolytic), homogenized, and kept for 15 min at room temperature. A 0.1-mL aliquot of this solution was diluted in 9.9 mL sterile saline and seeded with a 0.01-mL calibrated loop onto sheep blood and Baird–Parker agar plates. After incubation for 24 h at 37 °C, the number of colony-forming units (CFU) was determined by multiplying the number of colonies by the corresponding dilution [
36]. The cultures were classified as positive when the count was 10
6 CFU or higher. Microorganisms that grew in the culture medium were subjected to Gram staining for assessment of purity, observation of morphology, and specific staining. After confirmation of these features (Gram-positive cocci in clusters), the catalase and coagulase tube tests were performed according to Koneman et al. [
37] for phenotypic identification of
S. aureus. Catalase- and coagulase-positive isolates were subjected to DNA extraction (see item 4.6) for genotypic identification of
S. aureus with PCR amplification of the
Sa442 DNA fragment, which is specific for
S. aureus, following the protocol described by Martineau et al. [
38]. After the species was confirmed, the isolates were stored in nutrient broth with glycerol at −70 °C.
4.4. Antimicrobial Susceptibility Testing
The antimicrobial susceptibility/resistance profile of all
S. aureus strains isolated in this study was evaluated using the Kirby–Bauer disk diffusion method using disks impregnated with gentamicin (10 µg), linezolid (30 µg), penicillin G (10 IU), erythromycin (15 µg), and clindamycin (2 µg), as well as the D test for identifying inducible clindamycin resistance, according to the criteria of the Clinical and Laboratory Standards Institute [
39]. Methicillin susceptibility was determined using cefoxitin (30 µg) and oxacillin disks (1 µg).
4.5. Determination of Minimum Inhibitory Concentration (MIC)
The in vitro MIC of all
S. aureus isolates against oxacillin, vancomycin, linezolid, daptomycin, quinupristin/dalfopristin, and tigecycline was determined with the E-test. The MIC results are expressed as the proportion of isolates susceptible to each drug according to the CLSI definition [
39]. Isolates with intermediate values were classified as resistant.
4.6. DNA Extraction from S. aureus
Nucleic acid was extracted from all S. aureus strains isolated in the study. The isolates were cultured on blood agar, inoculated individually into BHI broth, and incubated for 24 h at 37 °C. Extraction was performed with the Illustra kit (GE Healthcare) according to the manufacturer’s instructions, and the extracted DNA was stored at −20 °C.
4.7. Detection of the mecA Gene
The parameters and primers described by Murakami et al. [
40] were used for the detection of the
mecA gene in the
S. aureus isolates. International reference strains were included in the assays as positive (
S. aureus ATCC 33591) and negative control (
S. aureus ATCC 25923).
4.8. Classification of SCCmec in S. aureus
All MRSA isolates were subjected to SCC
mec typing by multiplex PCR using the primers and parameters described by Milheiriço et al. [
36]. The following strains were used as a control for SCC
mec typing: COL for SCC
mec type I; N315 for SCC
mec type IA; PER34 for SCC
mec type II; AN546 for SCC
mec type III; HU25 for SCC
mec type IIIA; and MW2 for SCC
mec type IV.
4.9. Detection of Genes Encoding Virulence Factors
The genes for the following virulence factors were investigated in all
S. aureus isolates with PCR using the primers and parameters described in
Table 7: PVL (
lukS-
lukF-PV); toxic shock syndrome toxin (
tst); staphylococcal enterotoxins A to C (
sea,
seb, and
sec); biofilm (
icaA,
icaD, and
bap); exfoliative toxins A and B (
eta and
etb); and hemolysins α, β, and δ (
hla,
hlb, and
hld). International reference strains were included in all reactions as positive and negative control (
Table 7).
4.10. Visualization of Amplified Products
The efficiency of the amplifications was monitored with electrophoresis on 2% agarose gel prepared in 1X TBE buffer and stained with SYBR Safe. The size of the amplified products was compared to 100-bp molecular markers, and the gel was photographed under UV transillumination.
4.11. Characterization and Quantification of Phenol-Soluble Modulins (PSM) with High-Performance Liquid Chromatography (HPLC)
For the detection and quantification of PSM, 10 µL of an overnight culture was incubated in 1 mL tryptic soy broth for 16 to 18 h, centrifuged at 14,000 rpm for 15 min at 4 °C, and 300 µL of the supernatant was reserved for reversed-phase HPLC-mass spectrometry as described previously [
48]. PSMs were quantified by integrating the ion chromatograms extracted by the mass-to-charge ratios of doubly and triply charged ions of each PSM. The unit of PSM measurement is “amount arbitrary unit (A.U.)”.
4.12. Western Blot for the Quantification of spa and hla Expression
For analysis of protein A and alpha-hemolysin expression, the isolates were cultured overnight and 10 µL of this culture was added to 1 mL TSB and incubated for 8 h at 37 °C. The cultures were centrifuged, and a 300-µL aliquot of the supernatant was used for Western blot analysis. Supernatants were submitted at 95 °C for 5 min and loaded in the 12% SDS page gel. The applied voltage was 150 v, 400 mA for 1 h. The proteins were transferred to a nitrocellulose membrane, washed, and blocked with Rabbit antibody for alpha-toxin. The membranes were audiographed in the Typhon TRIO plus Variable Mode Imager®. GraphPad Prism® using a t-test analyzed the data.
4.13. Pulsed-Field Gel Electrophoresis (PFGE)
PFGE typing consisted of digestion with
SmaI of chromosomal DNA from
S. aureus isolates associated with VAP. The plug preparation protocol and parameters followed McDougal et al. [
49]. The restriction fragments resulting from
SmaI digestion were subjected to electrophoresis on 1% agarose gel in a CHEF-DR III system (Bio-Rad Laboratories, Hercules, CA, USA), with pulses alternating from 5 to 60 s at 6 V/cm and 13 °C for 23 h.
The gels were stained with Gel Red (Biotium, San Francisco, CA, USA) for 1 h, washed in Milli Q water for another hour, and photographed under UV transillumination. The profiles were analyzed with the BioNumerics software (Applied Maths, Sint-Martens-Latem, Belgium) using the Dice similarity coefficient and the UPGMA method (1.2 tolerance and 1% optimization) for cluster analysis of isolates. Clusters were defined when similarity was ≥80%. International clones were kindly provided by Antônio Carlos Campos Pignatari, Special Laboratory of Clinical Microbiology, Department of Infectology, Federal University of São Paulo-Escola Paulista de Medicina and Agnes Marie Sá Figueiredo, Paulo de Góes Institute of Microbiology, Federal University of Rio de Janeiro, Brazil.
4.14. Multilocus Sequence Typing (MLST)
The constitutive genes of
S. aureus associated with VAP were amplified and sequenced to determine allele composition and sequence type (ST) using the parameters proposed by Enright et al. [
50]. Seven housekeeping genes were used (
arcC,
aroE,
yqiL,
gmk,
tpi,
glp, and
pta). Sequencing was performed using Macrogen (Rockville, MD, USA), and sequences were analyzed using SeqBuider
® and the official website mlst.org.
4.15. Statistical Analysis
Poisson regression of host and microbiological factors was performed to evaluate their relationship with the development of VAP and death, using backward selection (the first model included all variables, and the multivariate model included variables with
p > 0.1, removing the variable with the highest
p-value at each step and including only those with
p < 0.1, i.e., significant, and marginally significant variables). The values obtained for the production of PSM, alpha-toxin, and protein A (spa) were dichotomized by the median for inclusion in the model (‘yes’ for a value equal to or greater than the median and ‘no’ for a lower value). Data were stored in Epi Info
® 3.5.2 (Centers for Disease Control and Prevention) and Excel and were analyzed using Epi Info and SPSS
®. The incidence density of
S. aureus and MRSA VAP was calculated using the following equation:
where ID = incidence density, NVAP = number of patients with VAP, 1000 = standard coefficient, ΣnVis = sum of the number of visits of patients (728), and 3.5 = interval between visits in days.