Streptomycetes are an excellent source of identified natural products due to their impressive ability to form a variety of interesting secondary products [1,2]. This has made these bacteria the focus of applications in industry and in research. These include applications in numerous fields in medicine, agriculture, and biotechnology. Streptomyces-derived natural product discoveries started in 1947 and peaked in the 1960s following a significant decline in the following decades. The development of combinatorial chemistry in combination with high-throughput screening and a rather limited funding for drug discovery contributed to this decline. Recently, new methods of genetic engineering, fermentation optimization protocols, and bioinformatics technologies including genome mining in addition to classic bioprospecting and bioassay-guided isolation reactivated the field. Different studies claim that the genus can synthesize some 150,000 more antimicrobial compounds than those currently known. These factors influenced recent constantly increasing interest in natural product discovery in the genus Streptomyces [3].

The secondary metabolites produced by Streptomycetes show structural diversity—the underlying chemical structure includes aminoglycosides, polyketides, non-ribosomal synthesized peptides, polypeptides, glycosides, terpenoids, lipoproteins, alkaloids, polyethers, lantibiotics, and other compounds. Examples of secondary metabolites include antibiotics (e.g., chloramphenicol, lincomycin, neomycin, streptomycin, tetracycline), immunosuppressants (FK-506, FK-520), antimycotics/fungicides (e.g., nystatin, natamycin, amphotericin B), herbicides (phosphinothricin), anti-tumor substances (doxorubicin), anthelmintic substances (avermectin), growth promoters in ruminant animal feed (monensin), anticholesterol and coccidiostatic substances (e.g., lasalocid), and insecticides (milbemycin) [2,4].

Despite the great diversity, many secondary metabolites in Streptomyces are based on similar biosynthetic mechanisms. Basic units and main stages are defined by a systematic regulated production process consisting of biosynthetic steps that can be found in different production strains. More than two-thirds of all known antibiotics are synthesized by streptomycetes [5], but there are many more unknown compounds produced by Streptomyces [6]. In the search for new natural products, different paths have been taken including biological and chemical screening as well as genome mining [7]. These strategies allow us to reveal the entire potential of Streptomyces including silent biosynthetic gene clusters [8]. Streptomycetes are not only the source of many bioactive compounds but are also important biotechnological producers of such substances. Polyketides are a particularly large and diverse group of secondary metabolites produced in complex biosynthesis.

Secondary metabolites are not essential for primary metabolism and homeostasis in Streptomycetes, but they protect cells from environmental stress and selective pressure [9]. For instance, siderophores serve to improve the absorption of iron from the environment [10], in which biosynthesis is induced by intracellular iron deficiency and can have a growth-promoting effect on the host organism [11,12]. Another example is the colored terpenoid carotenoids [13] produced to protect against photo-oxidative damage and oxygen radicals [14]. Their biosynthesis in Streptomycetes can be induced by light but is rather inconsistent [15]. Furthermore, pigments such as melanin can protect against UV damage [16,17]. Melanin has been shown to have antibiotic and antimicrobial activities and to act as a cation chelator and antioxidant [18]. Another important group of secondary metabolites from Streptomycetes is terpenoids that can act as antibiotics, hormones, odorants, and flavorings. For example, albaflavenone, a tricyclic sesquiterpene antibiotic [19], and ectoine are effective against osmotic stress [20] and are able to prevent protein misfolding [21,22]. In S. coelicolor A3(2), it was shown that hopanoids are formed during the transition from substrate hyphae to aerial hyphae [23]. Other secondary metabolites such as antibiotics also represent a fitness advantage in the fight for nutrients.

Antibiotics are low-molecular-weight metabolites (M < 2000 Da) with a diverse chemical structure derived from living microorganisms, usually at low concentrations (<200 μg/mL), in stepwise biosynthesis, and can inhibit the growth of other microorganisms [24,25]. Under laboratory conditions, the production of many antibiotics can be influenced by the medium or nutrient sources available from the medium. Diverse chemical compounds acting as activators for signaling cascades that promote the production of certain antibiotics [26] may induce cryptic gene clusters that are not activated under standard conditions.

The natural role of antibiotics can be studied only in habitat-like environments where they are in response to interactions with different organisms [27]. The actual concentrations of antibiotics are difficult to estimate in nature since their production depends, among other things, on the availability of nutrients [28]. Sites of action of antibiotics include essential processes such as nucleic acid synthesis, protein biosynthesis, cell membrane and cell wall-associated enzymes, and lipid acid biosynthesis of the cell [29,30], allowing killing or growth retardation of other microorganisms.

On the other hand, experiments with sub-inhibitory antibiotic concentrations (SICs), e.g., concentrations below the minimum inhibitory concentration (MIC), were shown to influence transcription, biofilm formation, and gene expression [31,32,33]. Secondary metabolites have also been described as auto-inducers of antibiotic production that can affect the biosynthesis of other antibiotics depending on the concentration of an inducer compound, which can be an antibiotic itself [34].

The biosynthesis of secondary metabolites is usually expensive to the cell, the costs of which are defined by the benefits of genes whose expression is often limited and subject to complex regulation. Diverse proteins are required for the biosynthesis of antibiotics, which are usually located in gene clusters with tightly regulated expression [35,36]. The number and organization of genes within biosynthetic gene clusters can vary greatly; mostly, these genes are required for backbone formation, modification, export, regulation, and resistance [37].

The regulation of secondary metabolism in Streptomycetes is very complex and is coordinated on different levels that are strictly coordinated with each other. The cell relies on the perception of growth-related signals that must be integrated into the complex regulatory network. The central components of the global as well as specific regulatory cascades are extra- and intracellular effector molecules. The transition from primary to secondary metabolism goes hand in hand with the morphological and physiological differentiation of Streptomycetes. On solid media, the onset of secondary metabolism is mostly associated with aerial mycelium formation involving the bld genes. In liquid culture, the start of secondary metabolism correlates with entry into the stationary growth phase, which is typically linked to nutrient limitation. For instance, it has been shown that low phosphate concentrations and a lack of amino acids that are sensed by the effector molecule ppGpp can act as a trigger for the production of secondary metabolites [38,39]. Furthermore, a relationship has been demonstrated between elevated cAMP (cyclic adenosine monophosphate) levels and increased production of secondary metabolites, particularly antibiotics but also other biologically active substances [40]. Another example is the γ-butyrolactones that represent a group of extracellular effector molecules in Streptomycetes [41] and other Actinomycetes [42]. They are involved in the regulation of secondary metabolism and morphological differentiation. In addition to global regulatory mechanisms, there are also biosynthetic cluster-specific regulators or regulations on systems. These are regulatory proteins that activate the transcription of genes of associated gene clusters, as well as proteins that repress the transcription of target genes. The underlying regulatory cascades can be very complex and interconnected [43].

In most cases, secondary metabolite biosynthetic genes are clustered, and all genes required for synthesis, export, or resistance are located next to each other in the genome [37]. To date, only a few exceptions have been reported, e.g., gene clusters for ansamitocin [44] and pristinamycin [45].

Until recently, most of the secondary metabolites in Streptomyces were discovered using bioassay-guided isolation and chemical characterization of compounds of interest. Because of the biochemical complexity of biosynthetic gene clusters (BGCs), the discovery of new secondary metabolites has been challenging. This limitation has been overcome recently with the development of a genome mining approach, in which advances in DNA and RNA sequencing technologies have resulted in a rapid increase in the number of high-quality Streptomyces genome sequences as well as transcriptomics data. The large number of genome sequences from this genome mining provides resources for novel secondary metabolite discovery using bioinformatic analysis of Streptomyces genomes in silico. Interestingly, recent reports have revealed that each sequenced Streptomyces genome contains approximately 20–50 BGCs, a greater number than previously known.

Most BGCs are not expressed or poorly expressed in Streptomyces under laboratory conditions (silent BGCs) [46]. Genomes of well-studied strains such as S. coelicolor, S. griseus, and S. avermitilis encode more than 30 BGCs; however, only three to five secondary metabolites were detected in these strains. For the activation of silent BGCs in Streptomyces strains, diverse methods have been applied, including heterologous gene expression, promoter replacement, overexpression, or repression of regulatory genes, refactoring of targeted BGCs, etc. A non-directed activation of silent BGCs has also been achieved using co-cultivation methods as well as the One Strain–Many Compounds (OSMAC) method [47,48,49].

For the identification of secondary metabolite BGCs from data obtained with full-genome sequencing, genome-mining tools have been developed. Some of the recently proposed bioinformatics tools include ClustSCAN, NP searcher, GNP/PRISM, and antiSMASH [46]. AntiSMASH is considered to be the most widely used genome mining pipeline that features a user-friendly web interface with the possibility to predict the broad spectrum of secondary metabolite BGCs [46] (Table 1).

For the selection of genes for precursor synthesis, a genome-wide in silico reconstruction of the metabolism (genome-scale metabolic networks) has been proposed [50,51,52]. It is possible to simulate the growth rate, production rate, and mutation of genes. Furthermore, BGCs may be identified using low-coverage sequencing of a plasmid library with a small insert size and subsequent database comparison. These short sequences then serve as a starting point for probe design and the screening of a cosmid or BAC library [53]. Reverse genetics is another approach for Streptomyces engineering. It is based on the fact that secondary metabolite BGCs contain conserved core domains. Using alignments of known proteins that have similar functions in the biosynthesis of secondary metabolites, it is possible to identify previously unknown genes from genome libraries. The availability of a large number of accessible secondary metabolite biosynthesis gene cluster sequences offers a good starting point to look for specific biosynthetic gene clusters [54].

Information about secondary metabolite BGCs obtained with genome mining is essential for secondary metabolite discovery. Furthermore, it is a resource for rational design facilitation of BGCs and yields improvement in compounds of interest. In particular, polyketides (PKs) and non-ribosomal peptides (NRPs) can be redesigned because these compounds are synthesized by connected modular enzymes, which are able to recognize module-specific amino acids or CoAs. An example is a successful replacement of module 7 of AveA3 and AveA1 in the BGC of avermectin of S. avermitilis with MilA1 and MilA3 in the biosynthetic gene cluster of S. hygroscopicus that led to milbemycin production in S. avermitilis [55].

For the activation of silent BGCs and triggering the production of a compound of interest, various strategies have been developed. One possibility is to induce BGC expression in a native Streptomyces host. An example of the application of this strategy is the titer optimization of KF-506 in S. tsukubaensis [56]. The advantage of this strategy is that such a host is genetically tractable and genetic manipulations of the host genome are possible, including the overexpression of regulatory genes and removal of competitive pathways with gene cluster knock-out, deactivation of negative regulatory genes, and replacement of native promoters with stronger promoters. Another possibility is to clone and/or refactor BGCs and transfer them into another non-native Streptomyces host for heterologous expression. This is useful for the activation of silent BGCs in genetically intractable Streptomyces sp. [46].

Classical tools for genetic manipulation of Streptomycetes include DNA overexpression, deletion, disruption, and replacement as well as the use of suicide and temperature-sensitive plasmids, which require selection and screening for single- and double-crossover recombination events (Table 1). These strategies are time-consuming and have comparably low efficiency—double-crossover mutants are rather rarely obtained in Streptomyces, which demonstrates inefficient DNA homologous recombination. To address these limitations, diverse genome editing technologies were introduced. On the one hand, techniques to express BGCs in a heterologous host have been optimized: acquisition of the target SM-BGC from the native host genome (e.g., using a genomic library of cosmids, fosmids, BAC, and PAC), ligation/assembly of the BGC to the vector (sticky/blunt end ligation, Gibson cloning, recombination in different hosts), transfer of the SM-BGC-encoded vector to the heterologous host for expression (conjugation, protoplast transformation), and target secondary metabolite production with the expression of an integrative or replicative BGC vector. On the other hand, new techniques have been proposed, including the PCR-targeting system [57], Cre-loxP recombination system [58], I-SceI promoted recombination system [59], SpCas9-based genome editing [60,61,62,63], CRISPRi-mediated gene repression for single cells [64], FnCpf1-based genome editing and CRISPRi [65], base editing tools [66,67], and alternative CRISPR/Cas-based genome editing [68]. Recent discoveries on CRISPR (the clustered regularly interspaced short palindromic repeat) and CRISPR-associated protein (Cas)-based tools further improved the genetic manipulation of Streptomyces sp., accelerating natural product discovery, strain improvement, and genome research [46]. Application of genetics parts, such as synthetic promoters (e.g., constitutive ermE, SF14P, kasOP, gapdh, and rpsL promoters as well as inducible tipA nitA and xylA promoters), ribosome-binding sites (AAAGGAGG and diverse native or synthetic RBSs), terminators (e.g., Fd, TD1) and reporter genes (e.g., genes luxAB, amy, xylE, and gusA as well as eGFP, sfGFP, mRFP, and mCherry) further expanded the toolbox for Streptomyces engineering [46] (Table 1).

For improvement in the yield of secondary metabolites, different genetically modified Streptomyces hosts (also referred to as “super-hosts”) were generated by removing endogenous BGCs as well as nonessential genes and genomic regions. These include optimized Streptomyces strains as heterologous expression hosts that were generated by removing BGCs resulting in strains that can conserve energy and SM building blocks and have a specific precursor pool. For example, engineered strains of S. coelicolor, S. lividans, S. albus, S. avermitilis, S. chattanoogensis, and multiple others demonstrated improved secondary metabolite production of target compounds and reduced background chemical profiles [69].

Streptomycetes possess mechanisms to control metabolic pathways, including the production of secondary metabolites, in response to external signals and nutrient availability. Antibiotic production can be induced by substrates for antibiotic-producing enzymes or by regulation of the biosynthesis, activity, and stability of these enzymes. Nitrogen-containing compounds were shown to indirectly regulate antibiotic production by affecting the primary metabolism that provides precursor molecules for secondary metabolite biosynthesis [70]. Feedback/feedforward regulation and the regulation of nutrient supply, especially in the production of antibiotics, have been demonstrated to be mechanisms that can lead to the enhancement and overproduction of secondary metabolites for industrial needs.