Background
Staphylococcus aureus is capable of causing a variety of human diseases in both hospital and community settings. Diseases caused by
S. aureus range from superficial skin infections to life threatening systemic bacteremia [
1]. The effectiveness of current antibiotic treatments that interfere with
S. aureus growth and viability is limited by the development of drug resistance. The rapid emergence of antibiotic resistant strains [
2,
3] highlights the urgent need for newer and safer strategies to combat infection. An alternative approach to overcome the pitfalls of drug resistance is to develop anti-infective agents that selectively disrupt virulence-mediated pathways without affecting microbial cell viability [
4‐
6] or by modulating the host natural immune defenses to combat the pathogen [
7].
A number of different strategies have been adopted to identify new anti-infectives and these include using whole-animal infection models such as
Caenorhabditis elegans[
8‐
12]. Whilst traditional
in vitro and whole cell drug screens are acknowledged as the established paradigm for identifying antimicrobial molecules, the whole animal approach allows for early and direct assessment of
in vivo drug efficacy, thus, eliminating compounds that are toxic to the host or with poor pharmacokinetic properties [
13]. As current understanding of host-pathogen interactions and bacterial pathogenesis continues to increase [
14,
15], the
C. elegans model presents an advantage in being able to simultaneously identify compounds that target bacterial virulence as well as host defense.
In a live animal model with a recognized host-pathogen interaction, potential hits that selectively disrupt the virulence pathways utilized by pathogens to establish infections can be identified [
13,
16]. Furthermore, several conserved innate immune signaling pathways have been revealed from studies of host-pathogen interactions using
C. elegans[
17]. Conceptually, anti-infectives that target nonessential genes are likely to impose a lower level of selective pressure and probability of resistance development [
18] whilst ensuring the preservation of host endogenous microflora.
Infection of
C. elegans by
S. aureus is a robust platform to elucidate the mechanisms of host-pathogen interaction [
19‐
22]. In this study, our aim was to extend these studies to the discovery of novel anti-infectives. To achieve this objective, we chose to establish a
C. elegans – S. aureus liquid-based screen to identify anti-infectives that extend the lifespan of
S. aureus - infected nematodes from a collection of locally acquired natural products and synthetic compounds.
Discussion
We have successfully established a liquid-based
C. elegans -
S. aureus anti-infective screen platform which is relatively easy to perform, time-efficient and easily adaptable for large-scale screening of compounds of interest. This simple live animal screen enabled us to identify 14 natural extracts (from eight different plants and one marine sample) and 14 synthetic compounds that significantly extended the survival of
S. aureus-infected worms. We also observed that five of the selected natural extracts conferred a survival advantage for nematodes infected by MRSA. In agreement with previously reported findings [
8,
9], our results demonstrate that this screen identifies not only substances with anti-bacterial properties, but also hits that do not affect bacterial replication
in vitro which are usually overlooked in traditional
in vitro cell-culture based screens.
We noticed the
S. aureus- infected worms under both treated and untreated conditions appear smaller in size, similar to previously reported observations on solid-medium [
22]. It is less likely that the small size of infected-worms is a result of starvation as shown by previous transcriptome analysis on fasting-response genes [
22]. The smaller size of the infected worms may be a consequence of energy conservation and delayed development to compensate for the effort in fighting off the infection. Unlike the uninfected population fed on
E. coli OP50, we observed that the surviving infected worms in the presence of most hits were also small in size and slow in development. As indicated in the antimicrobial tests, most of the positive hits did not interfere with the growth of bacteria at the concentration used in anti-infective screening (200 μg/mL). Due to the continuous exposure to pathogen in the screening medium, it is not surprising that the live worms resembled the phenotype of infected worms, instead of the phenotype of healthy worms fed on their natural food source.
In contrast to
Salmonella typhimurium that persistently colonizes and proliferates in the nematode gut [
39],
S. aureus causes transient infection and continuous exposure up to a certain period of time is necessary to achieve maximal worm killing.
S. aureus can grossly colonize and distend the worm intestinal lumen but does not persist within the infected host [
20]. Therefore, we modified the method of Moy et al. [
8] by exposing worms to live bacteria throughout the assay. A similar approach where the worms were not pre-infected has been used to screen a small molecule library for compounds that alleviate
P. aeruginosa-mediated killing [
11]. In addition, we were able to simultaneously evaluate compounds or extracts that inhibit bacterial growth and promote host survival by observing the turbidity of the liquid medium. This in turn, enables a quick selection of hits that do not affect bacterial proliferation yet enhance host survival, which will likely yield alternative therapeutics that circumvent bacterial resistance and potentially overcome the limitations of traditional antibiotics.
Among the 28 molecular entities that protected
C. elegans from lethal infection by
S. aureus, seven natural extracts and 13 synthetic compounds contain antibacterial properties against
S. aureus at different concentrations. As expected, both
C. longa and curcumin were able to inhibit
S. aureus growth as previously described [
40,
41]. On the other hand,
S. macrophylla leaves were previously reported to be active against
S. aureus[
42] but we show that the seed extracts did not affect
S. aureus growth. Notably, some extracts and compounds demonstrated antibacterial effect at very high concentration (MIC: 2000 μg/mL). Although they are potentially interesting as ‘probe’ molecules, these compounds have limited potential for commercial drug development in their current form.
Of all synthetic compounds tested, 24/29 contain a BHT moiety (Figure
1) in their chemical structure [
25]. Since the compounds tested are related to this structure, it is not surprising that the hit rate was high among the synthetic compounds. However, these hits do exhibit different activities suggesting that they may have multiple targets, some of which are not in common and to which they exhibit differential affinity. Up to 13/14 of the hits demonstrated either bacteriostatic or bactericidal effects against
S. aureus at different concentrations. BHT is a phenolic antioxidant which consists of a hydroxyl group (−OH) bonded directly to an aromatic hydrocarbon group. Other substances that exert comparable free radical scavenging anti-oxidant activity as BHT have been reported to have good antibacterial and antifungal activity [
43,
44]. Phenolic compounds have also been reported to exhibit antibacterial activity towards gram positive and gram negative bacteria [
45,
46]. The mechanism thought to be responsible for phenolic toxicity to bacteria includes forming hydrogen bonds with vital proteins such as microbial enzymes [
47]. Thus, the anti-
S. aureus property of all the synthetic compounds tested in this study may be associated with the presence of the phenolic moiety in the BHT chemical structure. Moreover, the backbone structures of the compounds (thiosemicarbazides, thiadiazoles, triazole-thiones and hydrazones) have been reported to possess antimicrobial activity against both gram positive and negative bacteria [
31,
32,
34]. This is believed to be able to elevate the antimicrobial properties of these compounds.
Among all synthetic compound hits, UC-10 appeared to be particularly interesting as it rescued the infected worms without affecting
S. aureus growth. UC-10 is a derivative of 4-(substituted benzylideneamino)-1,2,4-triazole-5-thione. In terms of structure, UC-10 differs from UC-11 and UC-12 at the second benzene ring, at which a –OH group was substituted at the ortho position. The different activity of this compound, as compared to the rest of the compounds, may be attributed to the presence of this additional –OH group and two butyl groups, resulting in two –OH groups flanking by tert-butyl groups into one structure (Figure
6). Modification of this aromatic ring diminished the antibacterial property of the 4-(substituted benzylideneamino)-1,2,4-triazole-5-thiones compound fused to the BHT moiety. Unexpectedly, this has resulted in its ability to rescue the infected host via a distinct mechanism, probably by targeting bacterial virulence or host immunity.
Some extracts and compounds were able to promote nematode survival at concentrations much lower than their respective MIC values. The identification of seven extracts and one compound that do not interfere with bacteria replication suggests that this group of substances may apply non-conventional strategies to permit host survival whilst not affecting cell viability. These compounds may attenuate bacterial factors essential for infection such as virulence determinants involved in host damage and disease [
4]. Additionally, some of these substances could modulate host responses towards the pathogen [
48] or act as an immunostimulator [
49] that induces the host immune defense to get rid of infection.
One of the most significant findings is the three different extracts from
S. macrophylla seed (UE-03-3, 03–5, 03–6) that prolonged the survival of both
S. aureus and MRSA-infected nematodes without diminishing the degree of
S. aureus accumulation in the
C. elegans gut. A similar effect was demonstrated in the study by Dharmalingam et al. [
24] where the ethyl acetate extract (UE-03-5) also protected
C. elegans from a lethal
P. aeruginosa infection without affecting bacterial viability and
in vivo proliferation of PA14 [
24]. The fact that this extract permitted the surviving host to harbor a substantial bacterial load in the gut following infection by
S. aureus and
P. aeruginosa indicates that this extract might have attenuated a bacterial virulence factor or pathway which is common among pathogens. Hence, the anti-infectives identified using the
C. elegans screen may be multifunctional and able to target a broad range of human pathogens. With the existence of universal virulence strategies employed by different pathogens and the conservation of innate immunity across phyla, anti-infective screening using the
C. elegans model has the potential to identify hits that are efficacious against different groups of pathogens.
Interestingly, extracts from
N. fruticans (UE-01-4) and
O. stamineus (UE-12), plants widely found in South East Asia, both demonstrated good anti-infective effects. Although the mangrove palm is well-known for its traditional use by the local practitioners to treat different ailments in Asia, the pharmacological effect of
N. fruticans remained poorly defined until 2011 when a group from Bangladesh reported that the leaf and stem of
N. fruticans exhibited anti-hyperglycemic and antinociceptive activities in a mouse model [
50]. On the other hand,
O. stamineus leaf extract is widely used as an herbal remedy for kidney and urinary tract disorders. Previous studies revealed the anti-oxidant [
51,
52], anti-apoptotic [
51], hepatoprotective [
52] and gastroprotective [
53] properties and also identified the major active fractions include eupatorin, sinensetin, 3′-hydroxy-5, 6, 7,4′-tetramethoxyflavone [
54] and rosmarinic acid [
55].
N. fruticans and
O. stamineus extracts enabled the host to survive both antibiotic-sensitive and antibiotic-resistant
S. aureus infection without interfering with bacterial viability, suggesting that these extracts possess compounds capable of targeting the host defense mechanism or attenuating bacterial virulence traits. The fact that the intestinal colonization of
S. aureus was blocked after exposure to these extracts further consolidates the possibility that these compounds enhance the host immunity to eliminate the pathogen or target the bacteria factor/s that prevent them from accumulating in the intestine . Therefore, extracts UE-01-4 and UE-12 are potential anti-infective candidates that may act by targeting the host immune defenses. Further investigations are ongoing to identify the active compound/s that actively promotes lifespan extension in
S. aureus-infected nematodes and the potential anti-infective underlying mechanism(s) of these candidates.
Not surprisingly, a small number of compounds accelerated the killing of nematodes by
S. aureus. When the worms were exposed to these substances and fed with heat-killed
E. coli OP50, a significant decrease in survival was observed as early as 24 hrs post-treatment compared to the untreated control implying that the substance is toxic to the host (Additional file
5). Hence, by using this
C. elegans screen, we are able to filter compounds that exhibit toxicity upon the host [
56].
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CK conceived and designed the experiments, performed the experiments, analyzed the data and prepared the manuscript including revisions. WAY and NAR were involved in synthesis of compounds and manuscript preparation. MWT and SN participated in the experimental design, data analysis and manuscript preparation. All the authors have read and approved the final manuscript.