1. Introduction
Staphylococcus aureus is widely recognized for its ability to adapt to different environments thanks to an excellent genome plasticity exerted through horizontal gene transfer (HGT) mechanisms [
1,
2,
3].
A large plethora of virulence factors and resistance genes makes
S. aureus an opportunistic pathogen that represented and still represents one of the most common causes of human infections [
4]. Emerged in the past as a nosocomial pathogen, Methicillin-Resistant
S. aureus (MRSA) has quickly spread in the community, following different evolutive paths through the sequential acquisition of mobile genetic elements (MGEs) [
5].
A major clinical issue related to
S. aureus is represented by persistent infections (e.g., prosthetic joint infections) characterized by an asymptomatic phase with relapses occurring months or even years after the apparent resolution of the infection [
6].
The ability of
S. aureus to evade the immune response through the activation of the “Immune Evasion Cluster” has extensively been assessed [
7,
8,
9]. Moreover,
S. aureus proved capable of surviving phagocytosis by neutrophils and macrophages, performing phagosomal escape, as well as of surviving inside non-phagocytic cells, such as osteoblasts, epithelial and endothelial cells [
10,
11], with all these mechanisms confirming its persistent behaviour. Intracellular survival may lead to the development of
S. aureus reservoirs in humans, contributing to both persistence and dissemination from the infection site to other body areas [
7,
9,
12].
In addition, these kinds of infections have often been associated with a particular
S. aureus phenotype called Small Colony Variant (SCV), an alternative bacterial lifestyle generally related to adaptation and persistence in unfavourable environmental conditions [
6]. Furthermore, the role of professional phagocytes as “trojan horses” has been largely observed in several infectious diseases, such as those caused by bacteria, fungi, and viruses. Interestingly, they are also involved in the carriage of pathogens across the Blood Brain Barrier (BBB) [
13,
14,
15,
16].
Understanding the intracellular survival mechanisms of different
S. aureus strains is crucial to developing effective strategies against difficult-to-treat infections. Based on our knowledge, intracellular survival of
S. aureus varies depending on the set of virulence factors produced (e.g., PVL toxin, alpha toxin, etc) as well as its genetic background. This finding was already assessed through internalization assays in human MG-63 osteoblast-like cells, which revealed differences in internalization and intracellular persistence of
S. aureus based on its sequence type (ST) [
17]. The complications related to the treatment of persistent and chronic infections has led to the development of novel therapeutic strategies. These include the use of natural antioxidant compounds to counteract bacterial infections by strengthening the host’s defense mechanisms. In this scenario, the natural dipeptide carnosine may represent a promising choice in addition to the antibiotics currently used [
18].
The aim of the study was to evaluate the ability of S. aureus USA300 to survive inside RAW 264.7 murine macrophages, as well as to examine potential genomic and transcriptomic changes involved in the transition from the wild type to the SCV phenotype in the presence of carnosine and erythromycin.
2. Results
2.1. Antimicrobial Susceptibility Test
Carnosine was tested alone in Cation Adjusted Muller Hinton Broth (CA-MHB) and Dulbecco’s Modified Eagle Medium (DMEM), and in a range of concentrations from 0,15 mg/L to 160 mg/L, showing no antimicrobial effects compared to the bacterial growth control.
S. aureus USA300 had an erythromycin MIC value of 32 mg/L in both CA-MH Broth and DMEM, also when tested in combination with carnosine 20 mM, showing no antagonism against the antibiotic.
2.2. Infection Assay and Evaluation of SCV Stability
The RAW 264.7 cell line was infected using a Multiplicity of Infection (MOI) of 50:1 (bacterial: eukaryotic cells) for 3, 24 and 48h. The osmotic lysis of infected macrophages was performed at every time point post infection (p.i), lysates were plated on Blood Agar (BA) and bacterial Colony Forming Unit (CFU) were counted after over-night incubation, to assess bacterial recovery compared to the starting
inoculum (2,5x10
6 bacterial cells) as well as to investigate the bacterial phenotype in all experimental conditions (
Figure 1).
As reported in Graph 1, a 2-log fold reduction was observed in the untreated control. Carnosine and erythromycin, both alone (20mM carnosine) and in combination (20mM carnosine and 32 mg/L erythromycin), yielded a recovery approximately 3-log lower than the starting
inoculum at each time point after infection. As expected, the SCV phenotype was also observed in the untreated control, confirming that
S. aureus tends to grow as SCV in order to counteract the cellular stress caused by macrophages. Even though SCVs were isolated in all experimental conditions, several subcultures showed different phenotypic stability based on the treatment. In particular, SCVs induced by either carnosine or erythromycin alone reverted to the wild type form right after the first passage on Blood Agar (BA) (
Figure 2a). On the other hand, SCVs isolated 48h p.i. from the combined treatment condition were stable for at least six passages on BA. This observation suggests that the combination of carnosine and erythromycin triggers and supports the mechanisms underlying SCV formation, leading to longer stability compared to other treatment conditions (
Figure 2b).
Graph 1. Colony count (CFU/ml) from all experimental conditions at 3, 24 and 48h p.i.
Graph 1. Colony count (CFU/ml) from all experimental conditions at 3, 24 and 48h p.i.
2.3. Genomic Analysis
Illumina Whole Genome Sequencing (WGS), de novo genome assembly, sorting and variants calling were performed. The analysis aimed to test crucial
S. aureus genes for one or more point-mutations which may represent a unique genomic signature of
S. aureus SCV. Genomic comparison did not reveal any relevant mutations which could differentiate
S. aureus USA300 wt and
S. aureus SCV. Indeed, all mutations observed are common to both phenotypes and not involved in aminoacidic changes that can affect protein structure and function. The detected mutations are listed in the supplementary materials (
Table S1).
2.4. Gene Expression Analysis
Gene expression was evaluated for genes belonging to several categories, so, regardless of their biological functions, they were classified as follows.
Metal Ions Uptake: up-regulation was observed for sirC, zur, mntR, fur, perR mRNA expression levels, with a p value <0.05 for fur, zur and mntR (S. aureus USA300 SCV vs S. aureus USA300 wt).
Ros neutralization: slight, statistically non-significant down-regulation and slight, statistically non-significant up-regulation were observed for katA and for sodM and sodA mRNA expression, respectively (S. aureus USA300 SCV Vs S. aureus USA300 wt).
Metabolic Pathways: slight, statistically non-significant down-regulation and statistically non-significant up-regulation were observed for pdhA and for fumC mRNA expression, respectively; on the other hand, statistically significant up-regulation was observed for uhpt, with a p value <0.05 (S. aureus USA300 SCV Vs S. aureus USA300 wt).
Regulators: slight, statistically non-significant up-regulation and remarkable, statistically non-significant up regulation were observed for sarA and sigB and for agrA mRNA expression, respectively.
Virulence: slight, statistically not significant down-regulation and remarkable, statistically non-significant up regulation were observed for psmA and for hla and hld mRNA expression, respectively.
Surface Protein: remarkable, statistically significant up-regulation was observed for
sdrE mRNA expression, with a p value <0.05 (
Figure 3,
Table 1,
Table S2).
3. Discussion
S. aureus is a notorious opportunistic pathogen that causes a plethora of diseases. As previously demonstrated, several
S. aureus strains exploit their ability to penetrate non-phagocytic cells inducing intracellular toxicity and tissue destruction; on the other hand, bacterial persistence within cells is thought to lead to immune evasion and chronicity of infections. Furthermore, phagocytosis by neutrophils and macrophages allows dissemination of the pathogen.
S. aureus host-pathogen interactions, with a focus on intracellular survival, is a matter deserving of serious consideration [
9]. Phagosomal escape is another mechanism of persistence in the human body that affects infection dynamics [
18].
Synthetized via the ATP-dependent enzyme carnosine synthetase 1 (CARNS1), carnosine is a dipeptide composed of β-alanine and L-histidine and found in several human tissues. The highest concentrations are observed in the skeletal and cardiac muscles (up to 20 mM) as well as in the brain (0.7–2.0 mM) [
19,
20,
21,
22]. Carnosine is extensively known for being involved in macrophage activation as well as a ROS and RNS scavenger, enhancing the expression of antioxidant enzymes (Gpx1, SOD-2, Cat) [
23].
Previous studies demonstrated a strong increase in IL-6 and TNF-α, two cytokines induced by
S. aureus ST239 when infecting non-immune cells such as osteoblasts; the pro-inflammatory cytokines produced during
S. aureus infections promote its outgrowth and bacterial intracellular persistence [
24].
In the present study, aimed at evaluating the effect of carnosine on macrophages and S. aureus persistence, the RAW 264.7 murine macrophages cell line was infected with the S. aureus USA300 strain and treated both with erythromycin alone and in combination with carnosine 20mM. Experimental conditions were evaluated at 3, 24 and 48 hours post infection. RAW 264.7 murine macrophages infected with S. aureus USA300 and not treated with antibiotics were used as controls, also demonstrating the great ability of this strain to survive in the cells. Small colony variants (SCVs) isolated from all analysed conditions demonstrate S. aureus USA300 adaptation.
Once SCVs were obtained from all experimental conditions, several passages on blood agar were performed in order to test their phenotypical stability. Despite a slight numeric decrement of the SCVs isolated 48 hours p.i. from the combined treatment with erythromycin and carnosine, a stable phenotype was observed for at least six passages on blood agar. These findings support the hypothesis that the use of carnosine as an adjuvant for erythromycin may support the macrophage antioxidant machinery triggering the host defence. On the other hand, this leads S. aureus to acquire a stable SCV phenotype.
As reported in the literature [
25,
26,
27], stable SCV phenotype is usually related to mutations in specific genes related to electron transport chain, such as
hemB,
menD,
ctaA and
thyA. Therefore, Whole Genome Sequencing (WGS) was performed in order to compare the wild-type isolate with its SCV. The analysis did not reveal any relevant mutation. Since chromosomal rearrangement might be involved in stable SCV formation [
28], the sequencing method employed may represent a limitation. Long-read sequencing is currently ongoing; this will provide greater genomic insights.
Gene expression analysis was performed with the aim of elucidating the differences between the wild-type and SCV phenotypes of
S. aureus USA 300, focusing our attention on target genes involved in regulation, adhesion and toxin production, and ROS neutralization. Since increased glycolytic metabolism was reported for stable SCV [
29], genes related to glycolysis were also investigated.
Figure 4 reports differentially expressed genes between
S. aureus USA300 and SCVs.
The expression of regulatory and virulence genes is reported in
Figure 3a,b and
Table 1, and summarized in
Figure 4.
The over-expression of the
agrA gene in SCVs may be explained by the over-expression of the
sarA gene; this expression profile is associated with intracellular survival in macrophages [
30].
SarA over-expression leads to the over-expression of
sdrE and virulence genes (
hla and
hld genes), thus favoring the mechanism of adhesion (a typical strategy in the initial stages of the internalization process, when
S. aureus shows a slowdown in metabolic activity) and persistence in macrophages [
31].
The expression of the Ros neutralization and Metal Ione uptake genes is reported in
Figure 3d,e and
Table 1, and summarized in
Figure 5.
Over-expression of the genes encoding for zinc, iron and manganese transporters (
zur,
sirC and
mntR) was reported, highlighting the importance of metal ions as cofactors of several enzymes involved in ROS neutralization, such as catalase and superoxide dismutase [
32,
33,
34] and their centrality for macrophage survival [
7,
30].
These regulators act on PerR, which affects the oxidative stress acting on catalase and superoxide dismutase encoded by
katA,
sodM,
sodA [
35,
36]. The latter was found to be over-expressed. These genes were involved in the detoxification of vacuolar reactive oxygen species, contributing to intracellular survival, which in this contest is not yet required [
7].
As previously demonstrated, the glycolytic and fermentative pathways are up-regulated in
S. aureus SCVs, even in the presence of oxygen, thus exhibiting a type of anaerobic metabolism [
37]. Indeed, statistically significant over-expression of the
uhpt gene (a hexose phosphate transporter) was observed. Moreover, statistically non-significant over-expression was shown for the
fumC gene. This gene encodes for fumarate dehydrogenase, responsible for the conversion of fumarate into malate. Fumarate is known to be essential for the epigenetic changes associated with trained immunity, therefore its depletion, due to
fumC up-regulation, was shown to be crucial for the intracellular survival of SCVs [
29].
Since chronic and persistent infections are a major clinical burden due to failure of prolonged antibiotic treatment, our findings allow to improve the knowledge of antibiotic resistance mechanisms, sometimes related to the most difficult-to-treat SCV phenotype.
In order to clarify the mechanisms involved in host-pathogen interaction, this experimental model should be improved to include the impairment of the host’s inflammation pathways that lead to the failure of macrophage immune activity.