Methicillin-resistant Staphylococcus aureus (MRSA) show resistance to almost all therapeutic β-lactams and other classes of antibiotics. MRSA was first reported in 1961 (Jevons, 1961), when a β-lactamase producing strain of S. aureus which had previously been methicillin sensitive, developed methicillin resistance. The fact that this occurred only a year after the introduction of the semi-synthetic penicillin was truly a harbinger of the specter of drug resistance that would haunt healthcare providers in the years to come. MRSA has since been isolated in many hospitals around the world and currently represents a serious healthcare problem. It is particularly prevalent (>50%) in South America, Romania, and Japan and is becoming increasingly widespread in other countries (Lee et al., 2018). Concerns have also been raised over the emergence of MRSA among livestock due to the extensive use of antibiotics to prevent and treat infections (Conceição et al., 2017). Recently, cases of MRSA have been reported outside of the hospital settings, mainly affecting young, healthy individuals (Braun et al., 2016; Braun and Kahanov, 2018).

While MRSA generally do not cause severe disease, there are limited therapeutic options available to MRSA infections making all infections, even mild ones, noteworthy. By definition, MRSA are resistant to penicillin-like antibiotics, and they have now been noted to be developing resistance to other existing classes of antibiotics (Kaur and Chate, 2015). There is a constant hunt for new antibiotics but in the last few decades, only a limited number have been added to the clinician's arsenal; among them are linezolid in 2000 (Lee and Caffrey, 2017), daptomycin (a lipopeptide) in 2003 (Frankenfeld et al., 2018) and ceftaroline in 2010 (Long et al., 2014). Presently, vancomycin remains the most important first-line therapy for severe MRSA infection. However, the emergence of MRSA with reduced susceptibility to vancomycin (Ghahremani et al., 2018) as well as daptomycin (Roch et al., 2017) and linezolid resistance (De Dios Caballero et al., 2015) have been reported. Given that bacteria naturally evolve toward developing resistance to all antibiotics they are exposed to, there is a critical need for research focusing on the search of novel antibacterial agents as well as innovative approaches to combat MRSA. In light of the pressing need for new anti-MRSA drugs, the World Health Organization has also included MRSA as an important antibiotic-resistant bacteria requiring the urgent need for new drugs (WHO, 2017).

Natural sources such as microbes, plants, and animals have contributed immensely to the development of current drugs (Gu et al., 2013; Tang et al., 2016; Ma et al., 2018; Tan et al., 2018). Among these natural sources, microbes, particularly those belonging to the gram-positive Actinobacteria phylum, stand out as a rich source of drugs (Bérdy, 2012). The genus Streptomyces is categorized under the phylum Actinobacteria (Waksman and Henrici, 1943); they currently represent the most widely studied genus under the Actinobacteria phylum with 843 species and 38 subspecies to date (LPSN, 2018). The vast diversity within this genus based on its sheer numbers is particularly evident when compared with other genera: Micromonospora genus has 84 species and 7 subspecies, Propionibacterium has only 16 species and 4 subspecies, while Salinispora has 3 species (LPSN, 2018) at the time of writing (June 2018). Based on historical evidence, Streptomyces seem to be a viable target in the hunt for new drugs as they represent the source of 75% of clinically useful antibiotics presently available (Janardhan et al., 2014). One of the newer antibiotics currently in use, daptomycin, represents the latest contribution of Streptomyces in the fight against pathogenic microbes—it was discovered in the 1980s and approved by the US Food and Drug Regulatory Administration (US FDA) for clinical use in 2003 (Frankenfeld et al., 2018). To date, Streptomyces-derived daptomycin remains the only naturally produced antibiotic of a novel class introduced since 2003 and it is currently considered a first-line drug for treatment of MRSA bacteremia (Choo and Chambers, 2016).

Streptomyces has a large genome, which logically contains many biosynthetic gene clusters (Bentley et al., 2002; Ikeda et al., 2003), a further indication of their potential ability to produce large numbers of compounds with diverse biological activities. However, under traditional culture condition, only a few compounds have so far been isolated—far less than what is expected based on the genome. Various methods discussed later are being used to overcome this problem. An additional recent concern is that analysis of Streptomyces from terrestrial soil has limited yield of new compounds but instead leads to rediscovery of known compounds. To overcome this problem, researchers now focus on isolating Streptomyces from underexplored ecosystems (Hong et al., 2009). Researchers are also attempting to utilize current genetic tools to identify gene clusters of promising compounds. Once identified, these biosynthetic gene clusters are modified in order to improve the efficacy of the compounds, produce better analogs of compounds, or increase the yield of the compound of interest (Alexander et al., 2010; Yang et al., 2017). Besides that, in situ computer-models have been used successfully to determine mechanism of action of some promising anti-MRSA compounds isolated from Streptomyces. Newer animal models have also been used to determine in vivo efficacy and toxicity of anti-MRSA compounds produced by Streptomyces. This review aims to highlight the potential of Streptomyces as a resource to combat MRSA—we look at all the anti-MRSA compounds derived from Streptomyces since the 1990s. We also discuss the ecological niches where the source organisms may be found, mechanism of actions of anti-MRSA compounds produced by Streptomyces and newer interventions for MRSA infection.

Hospital-acquired MRSA (HA-MRSA) is now among the most problematic bacterial infections to treat (Kaur and Chate, 2015) and, alarmingly, it is responsible for about 20–80% of hospital infections (Krishnamurthy et al., 2014). Even though the incidence of HA-MRSA is reported to have reduced by 54.2% in the USA (Dantes et al., 2013), statistics from other parts of the world indicate this is not the general trend. For example, studies in South Africa, India and Pakistan revealed 52, 54.8, and 50% of hospital-acquired infections, respectively, were attributed to MRSA (Laxminarayan et al., 2013). According to the European Centre for Disease Prevention and Control (ECDC), the number of cases of MRSA infection varies greatly between the northern and southern regions of Europe. For example, Norway and Iceland were reported to have the lowest case of MRSA infection of 1.2 and 1.3%, respectively. Romania and Portugal represent the opposite end of the spectrum with the highest rate of cases reported with figures of 50.5 and 43.6%, respectively, in regard to invasive infections (ECDC, 2016). Most often, in hospital settings, MRSA occurs as a secondary infection and is most prevalent among the elderly, post-surgical and immunocompromised patients (Krishnamurthy et al., 2014). These secondary infections lead to increased healthcare costs, resulting from prolonged hospital stay and additional antibiotics (Nelson et al., 2015). In Japan alone, a recent comprehensive comparative cost analysis for MRSA has been estimated to be greater than all other non-MRSA infection (Uematsu et al., 2017). As an indication of the scale of the threat posed by these organisms, in the USA, MRSA kills more people than HIV and TB combined (Boucher and Corey, 2008). In hospitals, controlling the spread of MRSA is further complicated by their ability to form biofilms on the surfaces of medical devices. These biofilms tend to be resistant to disinfectants and may act as reservoirs for growing MRSA colonies that can be transferred to another host (Suzuki et al., 2015). The carriage rate in hospitals is estimated to be around 20–60% (Pathare et al., 2016) with the resultant implication that a large proportion of health care workers (Shibabaw et al., 2013; Khanal et al., 2015; El Aila et al., 2017) and patients (Ho and Hong Kong intensive care unit antimicrobial resistance study (HK-ICARE) Group, 2003; Aslam et al., 2013; Moyo et al., 2017) are carriers of MRSA. The most common way MRSA enters a host is through a breach in the skin (Datta et al., 2014) and tends to develop into an infection when there is immunodeficiency in the host. As a result of its tendency to affect more vulnerable patient populations, MRSA has not only become a difficult disease to treat but also a costly one. While immunocompromised individuals are at higher risk of MRSA infection, worryingly there have been recent reports of MRSA infection among healthy individuals, especially children (Davoodabadi et al., 2016).

About 20 years after the first reported case of MRSA, the organism was confirmed to have spread beyond the hospital environment to the community. The earliest report of community acquired MRSA (CA-MRSA) was in Detroit, Michigan, USA in 1980 (Saravolatz et al., 1982). Several other community-based infections were reported not long after—in the community of native Indians (Taylor et al., 1990), then in 1989-1991 it was found among the Aborigines of Western Australia (Udo et al., 1993) and after that it was detected in Europe (Stegger et al., 2014). While HA-MRSA cases seem to be on the decline in the USA, CA-MRSA has emerged more strongly in communities around the world. Currently, 2 out of 100 people are carriers of CA-MRSA (CDC, 2016); which is particularly worrying as CA-MRSA can more easily spread than HA-MRSA. It has been suggested that there may be a more mobile genetic element in CA-MRSA as compared to HA-MRSA (Udo and Boswihi, 2017; Boswihi and Udo, 2018). According to CDC, a CA-MRSA is categorized as such if infection is evident on admission or a MRSA culture was obtained within 48 h of admission, with no history of admissions or medical treatment requiring invasive procedures (Gorwitz et al., 2006).

Carrier status is not limited to humans as there is evidence to suggest that animals have also become carriers of MRSA, thus creating the possibility of spread of MRSA from animals to humans. The steady increase in the global population has resulted in ever increasing demands for food supply. As part of the efforts to increase the yield of livestock production, antibiotics have been increasingly used to prevent infection. However, the uncontrolled use of antibiotics has encouraged the development of antibiotic resistance including the emergence of MRSA within these animals (Van Boeckel et al., 2015). These livestock (Conceição et al., 2017) and their products tend to become reservoirs of MRSA (Asiimwe et al., 2017). Cases of animal-associated MRSA transfer to humans have been reported (Loncaric et al., 2013; Van Duijkeren et al., 2015). A few cases where MRSA was recovered from free-living animals (Wardyn et al., 2012; Porrero et al., 2013) and pets (Bierowiec et al., 2016) have been reported. This additional reservoir of MRSA in animals in the community creates an additional threat to public health.

The notion that “there is a pill for every ill” has led to a widespread public perception that medication of some sort is necessary to cure all forms of illnesses. A particularly pertinent example is the use of antibiotics for virtually any infection—including mild bacterial infections that do not warrant antibiotic treatment and even viral infections. This practice is a major cause of the rapid rise of antibiotic resistant bacteria. As a countermeasure, many countries have imposed tight regulations on the use and sale of antibiotics; however even in these countries, prescribers were found to overprescribe antibiotics. Given this scenario, it is unsurprising that in countries with less stringent regulations, there is a tendency to abuse antibiotics. This overuse of antibiotics in humans and animals has accelerated development of antibiotic resistance (Ventola, 2015).

Antibiotic resistance in and of itself is actually a natural phenomenon forming part of bacteria's inbuilt machinery to help them to adapt to new and changing environments. Soil bacteria possess an inbuilt “resistome gene” that helps them express resistance mechanisms in response to external events (Nesme and Simonet, 2015). Resistome genes present in soil bacteria can be horizontally transferred to pathogenic bacteria over a period of time. In the presence of antibiotics, these organisms also tend to develop resistance through an antibiotic resistance gene (Nesme and Simonet, 2015). This demonstrates that antibiotic resistance is inherent in bacteria and underlines the need for constant research to help develop a new supply of effective antibiotics in order to treat infections, including MRSA that are resistant to almost all β-lactam antibiotics. However, in order to effectively develop new therapies, it is crucial to first understand the various mechanisms of drug resistance and the elements of the bacterial cells that are new potential targets for drug development.

Quorum sensing is the means through which bacteria sense the external environment in order to adapt to changes or stress and these may include pH, antibiotics, minerals Abisado et al., 2018; Igarashi et al., 2013) and even cell population density (Rutherford and Bassler, 2012). Regulating these changes helps bacteria to survive in critical conditions. The two-component signaling (TCS) system called histidine kinase sensor and its cognate response regulator (Utsumi, 2017) facilitate the regulation of changes that is required by bacteria in order to cope with the outside environment. To date, there are many TCSs found in bacterial populace (Fabret and Hoch, 1998; Lange et al., 1999; Kawada-Matsuo et al., 2013; Guo et al., 2017), of which however, the WalK/WalR system stand out as an important TCS for regulating cell wall metabolism (Zheng et al., 2015). Interestingly, it is so far found in gram-positive bacteria with low G + C content such as S. aureus. Among the 16 TCSs found in S. aureus, the WalK/WalR was observed to be single most important regulator for virulence and cell wall metabolism among others functions (Ji et al., 2016). In cell wall metabolism, the WalK/WalR system activates autolysins known as peptidoglycan hydrolases that facilitates restructuring of peptidoglycan layer of the cell wall and promotes continuous cell growth and division (Utsumi, 2017). Since the cell wall provides the structural support necessary for cell survival, the WalK/WalR system therefore plays a significant role in gram-positive bacteria such as MRSA. Further, WalK is a master regulator for cell wall metabolism as nine cell wall metabolism genes are dependent on WalK/WalR (Utsumi, 2017). Previous studies have shown that inhibiting the WalK/WalR system is detrimental and bactericidal to cells (Gotoh et al., 2010; Igarashi et al., 2013). Therefore, WalK/WalR system is a promising target for drug development of anti-MRSA therapy. Among other TCSs are those that regulate biofilm formation and virulence factors that may offer potential for future antibiotics to treat MRSA-related infections.

Streptomyces are filamentous gram-positive bacteria that are categorized under the phylum Actinobacteria (Waksman and Henrici, 1943). To date, there are about 843 species and 38 subspecies with validly published names in bacterio.net (LPSN, 2018) at the time of writing (June 2018). Streptomyces is known to be a very robust genus of bacteria in terms of their ability to thrive in inhospitable soil conditions. They have been isolated from soil samples from the richly biodiverse tropical regions to the far reaches of the Arctic Circle. The soil samples from these ecoregions seem to hold promising Streptomyces with interesting chemical and genetic make-up (Ser et al., 2015a, 2016b,c, 2017, 2018a) for drug discovery work. It was logical to begin a search for Streptomyces by investigating soil samples because soil is readily available and is known to be a rich source of microbes (Hibbing et al., 2010). Further, Streptomyces are saprophytic bacteria that thrives on dead and decaying materials. Most studies investigating the microbial diversity of soil samples have reported Actinobacteria, particularly Streptomyces as the predominant species. The fact that Streptomyces can be found in almost all places studied thus far, suggest that they are a highly competent bacterial species. On a molecular level, this demonstrates their superior genetic and metabolic potential that allows them to dominate the microbial population. To understand the mechanisms that led to the latter observation, researchers began studying related genes and proteins that could contribute to their versatility and ability to thrive. One of the interesting findings, which is now considered as a characteristic of Streptomyces, are their high guanine plus cytosine (G + C) genomic content. On average, they carry 70% G + C content which is considered very high as compared to 40% G + C content of Bacillus subtilis the model gram-positive bacteria (Kunst et al., 1997). In the most widely studied species S. coelicor, the G + C content measures up to 72.1% (Bentley et al., 2002). The high G + C content in Streptomyces species is believed to have accumulated over time by a selection process of adaptation to new environment; and is thought to have granted them overarching dominance in soil.

They also express powerful secretory systems constituting 40% of the ATP-binding cassette (ABC) transporters and MFS (Zhou et al., 2016). As a result, they have attracted interest from the biotechnology industry in the production of recombinant proteins. One example is the use of Streptomyces lividans as a potential producer for recombinant proteins due to the possession of better excretory system as compared to the traditionally used Escherichia coli (Anné et al., 2014). In terms of their ecological role, Streptomyces utilize these secretory systems, to regulate the intake of nutrients, expel toxins and secrete enzymes. These enzymes help break down tough plant and animal materials into soluble substances that can be easily taken up through their mycelia. Once in the mycelial cells, they are utilized for basic metabolic processes to produce ethanol, amino acids, nucleotides, organic acids and vitamins. Under stressful conditions, when soil nutrients become scarce, Streptomyces initiate a process called programmed cell death (PCD), whereby autolysin enzymes break down mycelial cells into amino acids and sugars, providing the building blocks for aerial hyphae that will eventually produce and carry spores. This process ensures that Streptomyces continue to grow, reproduce and survive under stressful conditions. A key point of interest from a drug development point of view is the biosynthesis of secondary metabolites that accompanies PCD. As the name suggests, secondary metabolites are derived from primary metabolites and cons diverse compounds with different biological activities (Ser et al., 2016b, 2018b; Tan et al., 2016). Some of these compounds are antibiotics and are today perceived to play an ecological role in terms of suppressing potential competitors by killing them or inhibiting their metabolic growth. This behavior becomes essential in protecting their food supply from other microbes and it demonstrates a metabolic pathway that is highly organized and well-co-ordinated, giving rise to their dominance (Ilic-Tomic et al., 2015). It is therefore not surprising to see that Streptomyces have a large genome size in order to facilitate production of the range of regulatory proteins and extracellular enzymes these processes require. Many of these secondary metabolites have been successfully used as antibiotics in treating infections in humans and animals (Cho et al., 2012), though resistance has now developed to many of them. The role of Streptomyces as an important antibiotic producer warrants their continued exploration as a promising source in the search for new anti-MRSA antibiotics.

Streptomyces first contributed to our antibiotic arsenal in 1940 with the discovery of the naturally produced antibiotic streptomycin from soil-derived Streptomyces griseus (Woodruff, 2014). Since then, 2 Nobel Prizes have been awarded to researchers for the study of Streptomyces (Woodruff, 2014; Dlugónska, 2015). Streptomyces have made invaluable contributions to conventional medicines (Janardhan et al., 2014) with 75% of antibiotics clinically used in humans having their origins in Streptomyces derived compounds while 60% have been used in animals (Cho et al., 2012; Ser et al., 2015b; Law et al., 2017a,b).

Although artificial synthesis of new drugs is theoretically promising, natural sources such as microbes remain the better producers of antibiotics as they provide lead molecules for development of current antibiotics (Ser et al., 2016a). According to the review of Bérdy, the success rate of drugs produced through chemical synthesis compared to microbes is 0.005 to 1.6 (Bérdy, 2012). Even today, microbe-derived chemical compounds continue to inspire the development of new antibiotics (Newman and Cragg, 2016). For example, some of the newly approved drugs from the year 2000 onwards are actually synthetic analogs of antibiotics produced by Streptomyces namely tigecycline, everolimus, miglustat, daptomycin, biapenem, ertapenem, pimecrolimos, and ceftarolinefosamil (De Lima Procópio et al., 2012; Kumar and Chopra, 2013). Similarly, Streptomyces are also the source of some drugs that are currently undergoing clinical trials using combination formulation of cephalosporin-lactam (e.g., ceftazidime-avibactam and ceftolozane-tazobactam), a new cephalosporin siderophore S649266, omadacycline (phase 2 trial) and eravacycline (Fernandes and Martens, 2017). Fosfomycin was previously isolated from S. fradiae in 1969 and is currently undergoing comparative phase 3 clinical trials to evaluate its intravenous preparation against piperacillin/ tazobactam in the treatment of chronic urinary tract infection and acute pyelonephritis in hospitalized adults (Fernandes and Martens, 2017).

Over the years, the number of compounds reported from Streptomyces have significantly reduced, resulting in fewer drugs approved for clinical uses. A clear example is the US FDA approved drug daptomycin-the only new class of antibiotic introduced since 2003. Overall, Streptomyces have produced an estimated 10, 400 bioactive compounds since the discovery of streptomycin. The number of compounds has waxed and waned over the years—bioactive compounds isolated from Streptomyces increased from 2,900 between 1940 and 1974 to 5,100 compounds in the years 1975–2000, before dropping to 2,400 compounds in the period between 2001 and 2010 (Bérdy, 2012). The decrease seen was mainly attributed to rediscovery of compounds from Streptomyces isolated from soil, particularly from the land (Bérdy, 2012). Given that genome mining has demonstrated the potential of Streptomyces to produce many more compounds than what is currently observed, the search for new bioactive compounds from Streptomyces continued with focus on other understudied ecosystem. Researchers began to focus on understudied environments particularly those with harsh condition postulating that the challenge to survival in these situations is the driving force for speciation resulting in a rich diversity of microbiological sources including Streptomyces (Hong et al., 2009). In the following paragraphs, we focus on the various ecological sources of Streptomyces with anti-MRSA activity.

Based on existing literature, Streptomyces derived from numerous ecological sources are active producers of natural compounds that exhibit anti-MRSA activity. These ecological sources include soil collected from terrestrial regions such as tropical forests, marine regions encompassing marine sediments and symbionts as well as newer understudied ecological niches such as endophytes, freshwater, deserts and mangrove ecosystem. A schematic diagram shown in Figure 1 is included to give an overview of the extent of work in the area of Streptomyces as potential sources of anti-MRSA agents.

Existing literature so far have highlighted 124 compounds produced by Streptomyces which show moderate to potent anti-MRSA activity (Supplementary Table 1). A numerical analysis of these compounds shows that polyketides (PKS) form the largest group (53), the next being non-ribosomal peptides (NRPS) while others include smaller proportions of alkaloids, hybrids of PKS/NRPS and PKS/terpenoids. In fact, polyketide and NRPS biosynthetic pathways are the source of most Streptomyces derived conventional antibiotics; and logically these classes also constitute a large fraction of the substances demonstrating anti-MRSA activity. Among the molecules identified were several compounds—such as polyketomycin, heliquinomycin, griseusin A, 4′deacetyl griseusin, citreamicin θ A, chaxamycin D, nosiheptide, nosokomycin, and marinopyrrole A—which have been reported to be potent anti-MRSA compounds and exhibit lower MIC than several clinically used antibiotics such as vancomycin. Vancomycin has been used as a standard for testing of effectiveness of new compounds against MRSA (CLSI, 2015). The chemical structures of these bioactive compounds are depicted in Figure 2A (1-12), Figure 2B (13-24).

An alternative to developing new anti-MRSA compounds is to focus on improving the effectiveness of currently available antibiotics such as β-lactams. Since developing a totally new drug is costly and time-consuming, a more cost-effective approach would be to enhance the anti-MRSA activity of existing drugs as measured by a significantly reduced MIC indicating the bacteria become increasingly susceptible to the levels of drug concentration in blood and tissues.

Antimicrobial drug combination approaches aimed at improving the effectiveness of anti-MRSA drugs (reducing MIC) has become a potential area of research in drug discovery. The urgent need to have new antibiotics to eradicate MRSA infection has further validated this area of research, especially when MRSA is resistant to most of the β-lactam antibiotics. Potentiating or enhancing the effectiveness of existing antibiotics through synergism studies has been used to identify new and better combinations with enhanced ability to eradicate infections. Synergism is a favorable effect observed in antimicrobial combination studies whereby activity is significantly greater when two agents are combined than that provided by the sum of each agent alone. Synergy is more likely to manifest when the ratio of the concentration of each antibiotic to the MIC of that antibiotic was same for all components of the mixture (Jain et al., 2011).

The rationale for combination of antibiotic therapy for bacterial infections in general is that it enhances activity, reduces toxicity, may prevent the emergence of resistance or even enable treatment of a polymicrobial infection. The checkerboard is a useful method for studying synergism since a number of combination concentrations can be prepared and studied via broth 2-fold dilution with synergism detected as no visible growth (Stein et al., 2015). According to the Clinical Laboratory Standards Institute guidelines for broth microdilution, the MIC was defined as the lowest concentration of antibiotic that completely inhibited the growth of the organism as detected with the naked eye (CLSI, 2015).

Biofilm formation is a key virulence determinant of MRSA; creating drugs able to induce disruption in the process of biofilm formation or its structure has become a potential area of research. This is because biofilm protects the MRSA from biocidals such as antibiotics and the immune system. Compounds from Streptomyces such as streptorubin B (20) have shown promising results in disrupting biofilm formation. Streptorubin B was originally isolated from Streptomyces sp. strain MC11024, and even though its MIC was 32 μg/mL, which is considerably higher than the vancomycin standard MIC of ≤ 2 μg/ml it demonstrated potential as a biofilm inhibitor by reducing the biofilm of MRSA to only 27% at 1 μg/mL (Suzuki et al., 2015). Since streptorubin B (20) was tested against one strain of MRSA, that is MRSA 315, increasing the number of strains tested will further demonstrate its inhibitory effect against MRSA biofilm. Streptorubin B (20) belongs to the previously discovered prodiginine class of pigmented antibiotics, known for their anti-tumor (Kojiri et al., 1993; Nakajima et al., 1993) and immunosuppressant actions (Songia et al., 1997) and now it has also shown potential as an anti-MRSA agent. Naturally occurring prodiginines have been isolated from actinomycetes, Serratia, Pseudomonas, and Vibrio among other bacterial species (Gerber and Lechevalier, 1976). From the limited data available, it seems justified to explore streptorubin B further to establish its role as an anti-MRSA agent.

A recent study showed that the ethyl acetate extract from Streptomyces sp. SBT343 significantly reduced the biofilm formation of several Staphylococcal species including USA300 (MRSA) at the maximum inhibitory concentration of 125 μg/mL. However, it did not display an inhibitory effect on the bacterial growth of Staphylococcus species as well as gram-negative Pseudomonas aeruginosa; overall indicating that it has selective anti-biofilm activity against Staphylococcal bacteria (Balasubramanian et al., 2017). Aside from high selectivity, no clinically relevant levels of cytotoxicity have been demonstrated, making this a promising source for an anti-biofilm against Staphylococcus species. Biofilms tend to complicate the therapeutic regime leading to increased duration of treatment, whereby the drug is required to penetrate through the biofilm to reach its target site. Most often, this can lead to treatment failure and consequently death. This has become a concern when treating severe MRSA infection with mainline drugs such as vancomycin. Using additional antibiotics with vancomycin, has been shown to dramatically improve the efficacy of vancomycin in both in vitro (Pistella et al., 2005) and in vivo mouse model (Shi et al., 2014). Given that vancomycin-intermediate and vancomycin-resistant S. aureus (VISA and VRSA) strains have been reported in some parts of the world, the need to develop new antibiotics is critical.

The in vivo stage of drug development involves the use of animals to investigate the safety and efficacy of new drugs in living mammalian organisms prior to clinical trials. This ensures that drugs are reasonably safe and effective before they are tested in human beings. However, prior to animal testing, promising drug candidates should, of course, demonstrate at least reasonable in vitro performance in terms of key parameters including MIC, rapid bactericidal activity and PAE (Guo et al., 2016).

These in vivo studies are necessary, because a compound that shows great potential in vitro, may not actually be effective in a biological system (Haste et al., 2011a,b). Unlike in vitro testing, in an actual biological system, a drug must first reach the target site where they can bind to and carry out their actions in order to have an effect; hence the pharmacokinetic processes of absorption, distribution, metabolism and excretion greatly influence their effectiveness and distinguishes the findings from in vitro and in vivo studies. Given that these drugs are usually intended ultimately for human use, it is important to select a model that mimics humans as closely as possible for the in vivo studies (Rosenthal and Brown, 2007).

Nosiheptide, which showed great promise in vitro, was tested in vivo on a murine model of intraperitoneal MRSA infection, where the test mice were given intraperitoneal injections of nosiheptide (20 mg/kg)—only 1 out of 10 infected mice died at the end of the study as compared to the controls where 6 out of 10 died on day 1. This provides evidence of significant in vivo activity of nosiheptide warranting further study (Haste et al., 2012). In vivo models are also used to determine toxicity of compounds isolated. Several of these compounds mentioned in Supplementary Table 1 are heliquinomycin, vinylamycin, polyketomycin, methylsulfomycin, lactinomycin, and gargantulide. Some of these compounds tested indicated toxicity in mice model, rendering them non-viable. Other examples include gargantulide isolated from marine Streptomyces which demonstrated a potent MIC of 2 μg/mL in vitro, methylsulfomycin which (at 25 mg/kg) was inconclusive due to precipitation formed while polyketomycin exhibited acute toxicity.

Moreover, the in vivo model selection needs to bear in mind several factors—mainly cost, time taken, and ethical considerations (Ferdowsian and Beck, 2011). Mammalian models tend to be more expensive and require a longer study period. Additionally, it is no longer possible to use mammalian models for most preliminary drug trials as the animal protection Act has tightened regulation of use of animal in experiments especially in European Union countries in 1998, where it is forbidden to use healthy animals in in vivo models (Uchida et al., 2014).

To overcome these problems, researchers have focused on non-mammalian animals such as Zebrafish, Caenorhabditis elegans, Drosophila melanogaster and silkworms as alternative hosts for in vivo screening systems. According to the literature reviewed in this paper (Supplementary Table 1), only silkworm and zebrafish have so far been employed for the screening of anti-MRSA compounds produced by Streptomyces. Uchida et al. (2014) demonstrated that a silkworm model could be successfully used in primary screening of new anti-MRSA agents—in this case, 4 potent anti-MRSA agents, nosokomycins A-D (MIC 0.125 μg/mL) (Supplementary Table 1) which were isolated from Streptomyces. Silkworm larvae infected with MRSA alone exhibited a lifespan of <3 days, compared to a survival rate of 3 days with vancomycin or other effective drugs including the nosokomycins (Uchida et al., 2014); the team also performed agar diffusion assay and also used a mouse model to provide supporting evidence of the reliability of the silkworm assay. The authors suggest that using a silkworm model as an in vivo-mimic assay may provide a more reliable prediction of actual in vivo activity compared to the traditional disk diffusion method. Hence, alternate animal models such as silkworms can become potential as alternate animal models in drug discovery and development of new anti-MRSA antibiotics.

Zebrafish embryos are an alternative model which is becoming increasingly popular in drug discovery work; they were initially introduced back in the 1970s but only recently became a model of choice in research on areas of genetically mutated diseases such as cancer. The advantages of this model are that they are cheap compared to rodents, they share 80% of genes with humans, and are much faster compared to trials on rodents. The zebrafish produce 200 eggs/week and so provides a reasonable source for research and they are easy to maintain. Additionally, the embryos are transparent making it easier to see the effects of test drugs on their organs and tissues. Zebrafish embryos were used to demonstrate the anti-MRSA activity of C23 (a purified compound from Streptomyces rubrolavendulae ICN3) as MRSA-infected zebrafish embryos survived with minimal toxicity effects in the presence of purified molecule C23 at 10 μg/ml (Kannan et al., 2014).

Streptomyces remain one of the most promising natural producers of antibiotics. Given the problems in rediscovery of known antibiotics from Streptomyces, it has been suggested that there may be a need to activate silent antibiotic gene clusters which may have gone dormant. This is possible with newer methods available such as: (1) The addition of signaling molecule N-acetylglucosamine in culture media of Streptomyces clavuligerus, Streptomyces collinus, Streptomyces griseus, Streptomyces hygroscopicus, and Streptomyces venezuelae have led to the activation of genes responsible for antibiotic productions. However, the same study also proved that this method does not work in all Streptomyces species (Rigali et al., 2008). In another study, Imai et al. (2015) showed that when Streptomyces are grown in the presence of lincomycin at sub-inhibitory concentration (1/12 or 1/3 its MIC) they produce compounds initially not produced in lincomycin-free media (Imai et al., 2015). (2) Similar results were seen with the addition of bacterial hormone gamma-butyrolactones (γ-butyrolactones), that serve as the main signaling molecules in Streptomyces species for the regulation of antibiotics by binding to cytoplasmic γ-butyrolactone receptors, whereby γ-butyrolactone induced gene expression of target genes and production of antibiotics in Streptomyces (Takano, 2006; Horinouchi and Beppu, 2007). Apart from γ-butyrolactones, the other 2 major groups of signaling hormones are the furans (e.g., 2-alkyl-4-hydroxymethylfuran-3-carboxylic acids) (Corre et al., 2008) and gamma-butenolides (γ-butenolides). In 2011, Kitani et al. (2011) was able to show that the γ-butenolide avenolide was responsible for stimulating the production of avermectin in Streptomyces avermitilis (Kitani et al., 2011). (3) Co-culture techniques are a powerful method of enhancing the biological activity of Streptomyces by allowing them to grow with another microbe in the same media (Sung et al., 2017). Sung et al. (2017) successfully used this method to enhance the anti-MRSA activity of extracts from marine Streptomyces sp. PTY087I2 M when grown in the presence of MRSA. The MIC was significantly reduced from 50 to 12.5 μg/mL, when grown under traditional culture condition and in co-culture system, respectively. These new approaches open possibilities to discover interesting anti-MRSA compounds from Streptomyces bacteria.

Also, understanding the biosynthetic pathways of important antibiotics is gaining increasing importance in the drug discovery process. A decade ago, cloning and identification of biosynthetic gene clusters was time-consuming and genome sequencing was expensive. Today, however, fast genome sequencing and genome mining have accelerated the identification of gene clusters–and knowledge of gene clusters provides crucial insights into modifying already existing antibiotics in order to improve them (Greule et al., 2017). These compounds may be effective in their ecological role to ward off competitors, but may need further modification in order to be viable antimicrobials in the human biological systems in terms of pharmacokinetics and dynamics. Apart from the new methods already mentioned in the paper having led to improved activity or increased yield, there are also reports of other approaches used recently to promote discovery of new compounds from Streptomyces. There are also reports of new methods of co-transformation carried out in Streptomyces sp. (Kallifidas et al., 2012). Despite the use of whole genome sequencing to identify the gene clusters, heterologous expression of these gene clusters often end up remaining silent. To overcome this problem, Kallifidas et al. (2012) used environmental DNA clones (eDNA), particularly those containing PKS type transcription factors and introduced them into S. albus harboring its corresponding minimal PKS containing clone. This method ultimately led to the production and isolation of tetarimycin in S. albus. In another example, Zhao et al. (2018) used a gene editing tool called the Clustered Regularly Interspaced Short Palindromic Repeats-Cas9 (CRISPR-Cas9) to introduce kasO promoter (97 base pair) successfully in Streptomyces roseosporus NRRL 15998. This promoter region then allowed for the activation of silent aurR1 gene cluster, which when expressed, led to the production of auroramycin (Zhao et al., 2018). This was only possible after the biosynthetic gene cluster of auroramycin (aurR1) was identified (Zhao et al., 2018). These methods can further improve already known anti-MRSA compounds from Streptomyces and also allow for discovery of unknown compounds that are currently not produced because of the fact that these genes remain silent under traditional culture condition.

Given that antibiotic resistance is an inherent property of bacteria, the need to develop newer anti-microbial agents over time is inevitable. However, it is vital to understand that abuse and misuse of antibiotics have greatly accelerated the development of antibiotic resistance. As such, stewardship programs play a key role in creating greater public awareness of the importance of compliance to antibiotics, consequences of overprescribing and sanitary practices. Efforts to reduce the development of antibiotic resistance will help preserve the effective lifespan of last resort antibiotics that are still effective against MRSA such as vancomycin.

In conclusion, MRSA poses a serious healthcare threat especially with the emergence of vancomycin resistant strains. Streptomyces have historically been and continue to play an important role as a source of new antibiotics—including compounds active against MRSA. Given the emerging trend of MRSA across the globe, the search for better therapeutic agents should be exploring all available resources—including Streptomyces from underexplored territory which are likely to yield exciting new compounds. This seems a promising approach since studies have shown that ecological variation provides a more suitable approach to accelerate the discovery rate of promising antibiotics. In this review, a number of new ecological niches have been discussed. Our review also revealed a spectrum of compounds isolated from Streptomyces with anti-MRSA activity and summarized the extent of studies so far carried out on these compounds.

Future work should focus on utilizing newly introduced genetic and chemical tools such as genome sequencing and mining and combinatorial biosynthesis to help create new and more effective antibiotics to address the emergence of antibiotic resistant infectious bacteria such as MRSA. Additionally, potential areas of research that remain yet to be undertaken have also been mentioned. The authors hope that future work can be focused on those underexplored niches, isolating new compounds and utilizing new technologies—genomic mining, combinatorial biosynthesis, mutasynthesis, to improve pharmacological properties of promising anti-MRSA compounds and thereby increase the number of successful anti-MRSA drugs for the treatment and halt spread of MRSA infection.

HK, LT-HT, TK, K-GC, PP, B-HG and L-HL performed the literature search, data analysis as well the manuscript writing. Technical supports and proofreading were contributed by PP, B-HG, and L-HL. L-HL and B-HG founded the research project.