Natural Products in Antibiotic Discovery

  • Fern R. McSorley
  • Jarrod W. Johnson
  • Gerard D. WrightEmail author
Part of the Emerging Infectious Diseases of the 21st Century book series (EIDC)


The development of antibiotics for human chemotherapy is one of the most important medical advances of the twentieth century; the vast majority of antibiotics in clinical use are microbial natural products or their semisynthetic derivatives. Pharmaceutical companies have been unsuccessful in identifying new antibiotics by screening libraries of synthetic compounds, because their chemical collections lack suitable chemical diversity tuned to entry into and retention by bacterial cells. Since natural products possess more of the physiochemical properties required for in vivo activity, there is a growing movement toward a return to these compounds for antibiotic discovery. Traditional screening methods for new antibiotics are plagued by the rediscovery of known compounds; consequently, new strategies are required to find new activity. For example, using medicinal chemistry to revisit discarded or underexplored scaffolds, or screening for adjuvants (e.g., inhibitors of resistance enzymes) can breathe new life into old antibiotics. The chemical diversity of antibiotics can also be increased by exploring new sources for antibiotic-producing organisms, by employing synthetic biology approaches using known scaffolds, and by mining genomes for silent or cryptic biosynthetic clusters.

17.1 History of Natural-Product Antibiotics

17.1.1 Natural Medicinal Therapies

Nature is rich in bioactive molecules that can be used as medicines. The earliest records describing natural medicines are found on clay tablets dating from 2600 BCE in Mesopotamia; these records contain over 1000 plant derived-substances [1, 2]. The most well-known ancient medicinal record is the Egyptian Ebers Papyrus, which dates from 1500 BCE and contains over 700 natural remedies, most of plant origin [3]. Natural/herbal treatments are found throughout history and from all over the world. Examples are the Chinese Materia Medica (Shennong Bencao Jing) (1100 BCE – 659 CE) [3, 4] and the Indian Ayurvedic system (1000 BCE) [3, 5]. These collections of ancient remedies were directed at a range of ailments that included infections; some are still in use today. Such traditional medicines generally consist of complex extracts and mixtures of agents whose bioactive component(s) went unidentified for hundreds of years. Improvements came largely from trial and error efforts that were hindered by confirmational bias and placebo effects.

In the early 1800s, morphine became the first bioactive natural product isolated from a medicinal plant [6]. This milestone led the Western medical field away from complex mixtures toward the pharmacology of pure compounds [7]. Old methods based on impure mixtures were unreliable: the variability of growth conditions and extractions from plant materials and microorganisms impacted the concentrations of both beneficial and toxic bioactive compounds present in the mixtures. Isolating the beneficial bioactive compound away from other material found in the natural source, some of which could cause negative effects, allows in-depth analysis of mode of action and efficacy. It also enables physicians to more accurately regulate the dosage of the bioactive compound.

In 1928, just over 100 years after the isolation of morphine, Fleming serendipitously discovered that the fungus Penicillium notatum secretes an agent that prevents the growth of bacteria. A decade and a half later, penicillin became the first natural-product antimicrobial to reach widespread clinical use [8]. Selman Waksman coined the term antibiotics to refer to microbially produced compounds that are “against life”; they either halt the growth of (bacteriostatic) or kill (bactericidal) microbes. Among these are small peptides having antibiotic activity [9]; they typically disrupt bacterial membranes in a nonselective manner. Indeed, these peptides are the first line of defense against bacterial infections. A larger suite of highly selective antibiotics is produced by microbes. These compounds have been the main source of our antibiotic drugs thus far [10]. The physiological roles of microbially produced antibiotics are still debated; these molecules evolved either for signaling functions or as chemical warfare agents to ward off neighboring microbes [11]. The density and diversity of resistance elements in microbes are consistent with an evolved detoxifying role to protect against the growth-impeding effects of antibiotics. Regardless of the evolutionary basis for antibiotics, the introduction of penicillin into the clinic in the early 1940s led to the “antibiotic era” of medicine . Natural antibiotics have aided in the treatment and control of infections for the last 80 years (Fig. 17.1). By controlling infections, antibiotics have revolutionized medicine, allowing physicians to perform lifesaving organ transplants and invasive surgeries and to treat cancer using disruptive immune system chemotherapy. The emergence of antibiotic resistance now threatens these breakthroughs, our quality of life, and our life expectancy.
Fig. 17.1

Timeline for the introduction of each class of antibiotic into clinical use. (Adapted from ref. [12])

17.1.2 The Development of Modern Antibiotics

The first antibacterial compounds to enter the clinic were synthetic molecules discovered in the laboratory of Paul Ehrlich [13, 14]. After noticing that some microbes stained differently than others when exposed to synthetic aniline and azo dyes, Ehrlich postulated that a chemical compound could selectively target pathogenic cells and not host cells. His “magic bullet” theory led him to screen hundreds of organoarsenic derivatives for activity. One of these, arsphenamine (Salvarsan), was successfully used to treat syphilis. Salvarsan, although difficult to administer, was the most prescribed antimicrobial drug until it was replaced by penicillin [15].

Four classes of synthetic antibiotics1 remain successful as drugs in the clinic today: sulfonamides, diaminopyrimidines, quinolones, and oxazolidinones (Table 17.1, Fig. 17.2). The initial class of systemic synthetic antimicrobials was the sulfonamide sulfa drugs [14, 17]. Prontosil, the first sulfa drug for human use, is a prodrug; after administration, it is metabolized into the bioactive agent sulfanilamide. Sulfonamides work by inhibiting dihydropteroate synthase, a critical component in folate synthesis. Humans acquire folate from their diet, while bacteria must biosynthesize this essential nutrient. Consequently, sulfa drugs are selectively active against microbes. Sulfonamides are generally co-administered with the diaminopyrimidine trimethoprim. This synthetic antibiotic also targets folate biosynthesis, inhibiting dihydrofolate reductase. The synergistic combination of trimethoprim and sulfamethoxazole (co-trimoxazole) is sold under a variety of trade names (e.g., Septra, Bactrim).
Table 17.1

Timeline of discovery and introduction of antibiotic classes

Antibiotic class; examples



Resistance observed

Mechanism of action

Activity; target species

Organoarsenics; salvarsan





Bactericidal; antisyphilitic

Sulfadrugsa; prontosil, sulfanilamide




Inhibits dihydropteroate synthetase in folate synthesis

Bacteriostatic; Gram-positive bacteria

β-Lactams; penicillins, cephalosporins, carbapenems




Inhibits penicillin-binding proteins in cell-wall biosynthesis

Bactericidal; broad-spectrum

Aminoglycosides; streptomycin, spectinomycin, kanamycin, neomycin




Binds 30S ribosomal subunit inhibiting protein synthesis

Bactericidal; broad-spectrum

Chloramphenicols; chloramphenicol




Binds 50S ribosomal subunit inhibiting protein synthesis

Bacteriostatic; broad-spectrum

Macrolides; erythromycin, clarithromycin




Binds peptidyl transferase center of 50S ribosomal subunit inhibiting protein synthesis

Bacteriostatic; broad-spectrum

Tetracyclines; chlortetracycline, doxycycline




Binds 30S ribosomal subunit inhibiting protein synthesis

Bacteriostatic; broad-spectrum

Ansamycins; rifamycins rifampicin




Binds RNA polymerase β-subunit inhibiting RNA synthesis

Bactericidal; Gram-positive bacteria

Glycopeptides; vancomycin, teicoplanin




Transpeptidase blockade inhibiting cell-wall biosynthesis

Bactericidal; Gram-positive bacteria

Quinolonesa; ciprofloxacin




Binds DNA gyrase inhibiting DNA synthesis

Bactericidal; broad-spectrum

Streptogramins; pristinamycins




Binds 50S ribosomal subunit inhibiting protein synthesis

Bactericidal, Gram-positive bacteria

Oxazolidinones; linezolid




Binds peptidyl transferase center of 50S ribosomal subunit inhibiting protein synthesis

Bacteriostatic; Gram-positive bacteria

Lipopeptides; daptomycin




Depolarization of cell membrane

Bactericidal; Gram-positive bacteria

Mutulins; pleuromutilin, retapamulin




Binds peptidyl transferase center of 50S ribosomal subunit inhibiting protein synthesis

Bacteriostatic; Gram-positive bacteria

Fidaxomicin (targeting Clostridium difficile)




Inhibition of RNA polymerase

Bactericidal; Gram-positive bacteria

Diarylquinolinesa; bedaquiline




Inhibition of F1FO-ATPase

Narrow-spectrum activity, Mycobacterium tuberculosis

aSynthetic antibiotic classes. Table assembled from references [11, 12, 16]

Fig. 17.2

Synthetic antibiotics used in the clinic

The quinolone drugs target type II DNA topoisomerases and replication. These agents are potent against both Gram-negative and Gram-positive bacteria. A widely used example is ciprofloxacin (Fig. 17.2), a second-generation fluoroquinolone that is orally available and used to treat urinary tract infections, sinusitis, and many other infections. Several newer generations of quinolones have found a place in the market, making the fluoroquinolones the most successful class of synthetic drugs to date.

The oxazolidinones comprise the fourth class of synthetic antibiotic. These compounds, which block the ribosomal peptidyl transferase center essential for protein synthesis, are effective primarily against Gram-positive bacteria. Linezolid (Fig. 17.2) represents the first generation of this class; it was approved by the US Food and Drug Administration in 2000 and became the first novel chemical scaffold to enter the clinic in several decades.

Fleming’s discovery that P. notatum secretes a bactericidal substance helped launch the intensive mining of microbes as sources of antibiotics rather than examining synthetic chemical libraries [8]. Florey, Chain, and Heatley’s efforts to develop a protocol for isolation of penicillin, followed by in vivo efficacy studies in animals and clinical tests in humans, showed that penicillin, a natural product, was a viable antibiotic drug candidate. The discovery of penicillin gave rise to the “golden age” of antibiotic discovery (Fig. 17.1). For the next 20 years, extensive screening of microbes, particularly soil-dwelling actinomycetes, was conducted to identify natural antibiotic compounds. In that time many natural-product scaffolds (aminoglycosides, chloramphenicol, macrolides, tetracyclines, ansamycins, glycopeptides, and streptogramins) were identified and often rapidly developed for clinical use (Table 17.1). By the middle of the 1960s, this simple but effective screening platform [18] appeared to have exhausted its sources, as new antibiotic scaffolds suitable for drug development became harder and harder to find.

The decline of success in isolating new chemical scaffolds from natural sources suitable for drug development, the generally poor pharmacological and toxicological properties of natural products as drugs, and the problematic emergence of resistance ushered in the next phase of antibiotic innovation. The focus shifted to medicinal chemistry efforts as we entered the “medicinal chemistry era.” Synthetic chemists prepared and derivatized core antibiotic scaffolds already used in the clinic and screened them for improvements. Chemists were successful in creating so-called “generations” of enhanced synthetic variants that improved pharmacological properties, expanded antibiotic spectra, and evaded resistance. Although many improved drugs and new generations of known drugs emerged from these efforts, no truly novel chemical scaffold entered the clinic from the 1960s to 2000s.

The lack of innovation in antibiotic discovery over the past two decades and the general failure of in vitro target-based drug discovery methods have prompted a renewed interest in natural products [16, 19]. This return to the natural-product compounds that previously dominated antibiotic drug discovery reflects the historic success of these drugs, a growing understanding of the physiochemical properties of small molecules for efficacy against bacterial targets, and new thinking resulting from advances in bacterial genomics, synthetic biology, and the properties of antimicrobial targets.

17.2 Major Classes of Natural-Product Antibiotics and Their Modes of Action

Microbial natural products are the source of most of our antibiotic scaffolds in current clinical use. Brief descriptions of the most prominent, clinically used drugs and their modes of action are outlined below.


Penicillin falls into the β-lactam category of natural products. Five classes of β-lactams are important in the antibiotic field: penams, cephems, carbapenems, clavams, and monobactams (Fig. 17.3). All β-lactams contain a strained 4-membered β-lactam ring system that in the majority of clinically relevant compounds is fused to a 5- or 6-membered ring system.
Fig. 17.3

β-Lactam antibiotics. (A) Major structural classes of β-lactam antibiotics. (B) The deacylation of penicillin G to generate 6-APA opened the door for the synthesis of new semisynthetic penicillins having improved antibiotic potency and spectrum of activity. (C) Similarly, 7-ACA has been used as a key intermediate for the preparation of countless semisynthetic cephalosporins having potent, broad-spectrum activity against clinically important Gram-negative bacteria. Ceftaroline has activity against methicillin-resistant Staphylococcus aureus (MRSA). (D) Natural and synthetic carbapenem antibiotics. (E) Aztreonam, a representative monobactam

Penams, cephems, and carbapenems covalently bind to penicillin-binding proteins (PBPs), essential enzymes that process the D-Ala-D-Ala termini of the pentapeptide portion of the peptidoglycan intermediates in cell-wall metabolism. The electrophilic β-lactam antibiotics mimic the D-Ala-D-Ala substrate (Fig. 17.4) and covalently bind to the PBP [20], preventing the enzyme from facilitating transpeptidation in the final step of peptidoglycan synthesis. This results in a complex series of molecular events that include inhibition of cell division and eventually cell rupture.
Fig. 17.4

β-Lactams mimic the D-Ala-D-Ala terminus of the peptidoglycan peptide strands and block the cross-linking step in cell-wall biosynthesis. (A) Structural analogy of penicillin with the D-Ala-D-Ala portion of peptidoglycan, as proposed by Tipper and Strominger in 1965 [20]. (B) Interactions of a penicillin with penicillin-binding proteins (PBPs) and β-lactamases. The acylation of PBP active-site serine prevents cross-linking of the bacterial cell wall and leads to cell lysis. In contrast, β-Lactamases hydrolyze β-lactams rapidly and confer high-level resistance. (C) β-Lactamase inhibitors that are used in combination with β-lactams to overcome resistance due to β-lactamases

Thousands of penicillins and cephalosporins have been synthesized through the acylation of 6-aminopenicillanic acid (6-APA) and 7-aminocephalosporanic acid (7-ACA) in order to improve antibiotic potency, the antibiotic spectrum against both Gram-positive and Gram-negative bacteria, and their stability against β-lactamases (Fig. 17.3b) [21]. β-Lactamases are resistance enzymes that rapidly cleave the β-lactam ring and render the antibiotic inactive. Although clavulanic acid and penam sulfones were found to be poor antibiotics, they are potent inhibitors of β-lactamases. Thus, they are combined with β-lactam antibiotics to successfully overcome resistance due to β-lactamases. Augmentin (amoxicillin + clavulanic acid), Unasyn (ampicillin + sulbactam), and Zosyn (piperacillin + tazobactam) are among the most successful combinations [22]. Two other β-lactamase inhibitors, avibactam and vaborbactam (RPX 7009), have been developed and approved recently for use in combinations with β-lactams [23, 24].

The discovery of several monobactams produced by soil-derived Gram-negative bacteria inspired the development of antibiotics based on a more simplified 4-membered lactam ring [25]. Before this discovery chemists had not considered using an N-sulfonic acid substituent to stabilize the β-lactam system. This modification eventually led to the successful antibiotic drug aztreonam (Fig. 17.3e) [26]. This example illustrates the importance of natural-product discovery in guiding the synthesis of new scaffolds.


In 1944, Waksman’s laboratory discovered streptomycin, the first aminoglycoside antibiotic , as a product of a strain of Streptomyces griseus [27]. Over the next two decades, several members of the class were discovered that included kanamycin in 1956 from Streptomyces kanamyceticus and gentamicin C in 1963 from Micromonospora purpurea [28, 29]. These were the first antibiotics to show efficacy in the treatment of tuberculosis and in infections caused by Gram-negative bacteria. Aminoglycosides consist of a core aminocyclitol ring modified to varying extents by saccharides. Multiple amino groups provide a positive charge at physiological pH and impart high water solubility. Currently, three aminoglycosides are commonly used in the clinic: the natural products, gentamicin C and tobramycin, and the semisynthetic agent amikacin, each of which contains the 2-deoxystreptamine aminocyclitol core (Fig. 17.5a) [30]. Streptomycin, which is unique in this drug class for having a streptamine core aminocyclitol ring, continues to find some clinical use in the treatment of tuberculosis and in combination with penicillins for enterococcal infections that are difficult to treat.
Fig. 17.5

Aminoglycosides. (A) Structures of natural and semisynthetic aminoglycosides. (B) Neomycin B and its binding mode in the H44 region of the E. coli ribosome (PDB: 4V9C). Asterisks at G-1405 and A-1408 indicate sites of methylation by methyltransferases that confer aminoglycoside resistance (e.g., ArmA and NpmA, respectively). (C) Major target sites of aminoglycoside-modifying enzymes (AMEs), including O-phosphotransferases (APHs), N-acetyltransferases (AACs), and O-nucleotidyltransferases (ANTs)

The amino and hydroxy groups of the aminoglycosides interact with the 16S rRNA in the 30S ribosomal unit through a network of hydrogen bonds (Fig. 17.5b). The binding of aminoglycosides to the 16S rRNA results in a conformational change that impedes cognate codon–anticodon validation by the ribosome. The result is the corruption of the genetic code and the synthesis of proteins with incorrect amino acids. This corruption contributes to cell death.

Resistance to aminoglycosides can take many forms. Active efflux can be a significant contributor to resistance in bacteria such as P. aeruginosa; however, the main mechanisms of resistance are chemical modification of the drugs or the target [30]. A large number of aminoglycoside N-acetyltransferases, O-phosphotransferases, and O-adenylyltransferases are present in both Gram-positive and Gram-negative pathogens (Fig. 17.5c). Over the past decade, ribosomal methyltransferases that modify the 16S rRNA (e.g., G-1405 and A-1408) and confer high-level aminoglycoside resistance in Gram-negative pathogens have also emerged as significant clinical challenges (Fig. 17.5b).


The term macrolide is a portmanteau that combines macrolactone, a lactone ring containing eight or more atoms, and polyketide. Erythromycin is a first-generation macrolide that was isolated from a soil actinomycete Saccharopolyspora erythraea. Erythromycin contains a 14-membered macrolactone framework and a 2-amino sugar (Fig. 17.6). The presence of a ketone at position 9 of the macrolactone can result in formation of a hemiketal with the hydroxyl group at position 6 under acidic conditions (e.g., exposure to gastric acids), thereby decreasing the levels of bioavailable drug [31]. Semisynthetic conversion of erythromycin to clarithromycin or azithromycin removes this possibility and results in improved efficacy and bioavailability. Macrolide antibiotics are most effective against Gram-positive bacteria, but they also have efficacy against some common Gram-negative, upper respiratory tract pathogens.
Fig. 17.6

Selected examples of macrolides (A), tetracyclines (B), and ansamycin antibiotics (C)

Macrolides interact with the peptidyl transferase center of the 50S ribosomal subunit, blocking peptide-chain elongation. Specifically, the hydrophobic surface of the macrolide binds to the sidewall of the exit tunnel and causes premature release of short peptidyl-tRNAs [31].

The main causes of clinical macrolide resistance are the 23S rRNA methyltransferases (Erm) [32]. However, a number of macrolide kinases, which modify the desosamine sugar, are increasing in prevalence and diversity. Ring-opening esterases and desosamine glycosyltransferases are also known, but they are less frequently encountered. Active macrolide efflux is common in many Gram-positive pathogens.


Chlortetracycline (Aureomycin) and oxytetracycline (Terramycin), discovered in 1948 and 1950, respectively, were the first members of the tetracycline class (Fig. 17.6b). Tetracyclines contain a tetracyclic polyketide core of four fused 6-membered rings; they were the first broad-spectrum antibiotics (i.e., effective against both Gram-positive and Gram-negative pathogens) to enter clinical use. After it was found that the C6-hydroxy group could be reductively removed to form a more stable 6-deoxytetracline, further modification ensued. Multiple generations of tetracyclines have now emerged that include doxycycline, minocycline, and tigecycline [33]. Tigecycline (approved for use in the USA in 2005) has broad-spectrum activity for both Gram-positive and Gram-negative bacteria; it is also effective against methicillin-resistant Staphylococcus aureus (MRSA).

Before the emergence of resistance limited their use, the tetracyclines were used for decades to treat infections of the respiratory tract, middle ear, and urinary tract. Like aminoglycosides, tetracyclines bind to the 30S ribosomal subunit. However, they do not cause the production of aberrant proteins. Instead they bind to the aminoacyl-tRNA binding site of the ribosome, thereby competitively preventing translation of mRNA.

Tetracycline resistance in Gram-negative bacteria is most often the result of active efflux. In Gram-positive bacteria, the expression of ribosomal protection proteins lowers the affinity of tetracycline for the bacterial ribosome. Enzyme-mediated inactivation has been reported through the action of TetX, a flavin-dependent monooxygenase that hydroxylates the antibiotic, thereby precipitating a nonenzymatic breakdown of the compounds [34].


As with the macrolides, ansamycins are polyketide macrocycles; however, ansamycins cyclize to form a macrolactam instead of a macrolactone (Fig. 17.6c). The ansa-bridged macrolactam is formed with an intramolecular amine nucleophile derived from the biosynthetic starter unit, 3-amino-5-hydroxybenzoyl-CoA. The natural ansamycin, rifamycin, is an 18-membered macrolactam that is converted through semisynthesis to the commonly used rifampin. Addition of the piperazinyl hydrazide to the rifamycin naphthyl core in rifampin increases its oral bioavailability and broadens antimicrobial activity. Rifampin is a WHO essential medicine designated for use in combination therapy for the treatment of tuberculosis.

Rifamycins are inhibitors of transcription [35]. In particular they bind to the mRNA exit site of the β-subunit of RNA polymerase. Resistance mutations in the rifamycin-binding site are common. As a result, rifamycins are most effective when in combination with other antimicrobials [35]. In the treatment of M. tuberculosis, which is the main use of the rifamycins, resistance is most often the result of point mutations in the target RNA polymerase [36]. In contrast, many environmental microbes express a wide variety of rifamycin-inactivating enzymes (e.g., kinases, ADP-ribosyltransferases, monooxygenases, and glycosyltransferases).


As their name implies, glycopeptides are peptides that are decorated with sugar moieties. Vancomycin and teicoplanin (Fig. 17.7) exemplify this class; both have been developed as Gram-positive-directed antibiotics. Vancomycin, which was discovered in the early 1950s as a product of the actinomycete Amycolatopsis orientalis, was used only sporadically for several decades, largely due to difficulties in obtaining pure compound. However, use increased in the 1980s following the widespread emergence of MRSA in hospitals [38]. Emergence of resistance in enterococci (VRE) and then intermediate resistance in S. aureus (VISA) spurred the development of second-generation glycopeptides such as telavancin, dalbavancin, and oritavancin that are less susceptible to resistance [39].
Fig. 17.7

Glycopeptide antibiotics and the interaction between vancomycin and the D-Ala-D-Ala portion of lipid II. In vancomycin-resistant enterococci (VRE), the peptide stem contains a D-Ala-D-Lac terminal region, and the affinity for the glycopeptide is decreased 1000-fold [37]

Vancomycin and teicoplanin are highly cross-linked pentapeptides that have a high affinity for D-Ala-D-Ala termini of uncrosslinked peptidoglycan chains. Vancomycin forms five hydrogen bonds with the D-Ala-D-Ala terminus of lipid II and prevents the formation of interpeptidyl cross-links by PBPs. That reduces the integrity of the cell wall and leads to cell death. Although glycopeptides and β-lactams both inhibit cell-wall biosynthesis, the glycopeptides sequester the substrate of transpeptidation rather than directly interacting with the PBP catalyst.

Resistance to glycopeptide antibiotics can take two forms. In Enterococci, reprogramming of cell-wall biosynthesis to terminate in either D-Ala-D-Lac or D-Ala-D-Ser reduces affinity of the antibiotic. In Staphylococcus aureus, acquisition of the genes encoding D-Ala-D-Lactate biosynthesis is known but exceedingly rare. Instead, in this organism glycopeptide resistance is primarily the result of increased production of cell-wall polymers that bind the antibiotic.


Dalfopristin and quinupristin (Fig. 17.8a) are semisynthetic derivatives of virginiamycins/pristinamycins that belong to the A-type and B-type streptogramin families. Type B streptogramins are cyclic hexapeptides, while type A streptogramins are cyclic polyketide–peptide hybrids. This illustrates that a pair of molecules, which have only bacteriostatic activity and are not effective treatments alone, can be combined to work synergistically and form a potent bactericidal drug. The molecular mechanism of synergy is based on the affinity of the compounds for different but adjacent regions of the bacterial ribosome [40]. Dalfopristin binds to the peptidyl transferase center where it reduces the affinity of aminoacyl-tRNAs for the aminoacyl site, which lowers subsequent peptide bond formation and chain elongation in the peptidyl site. In contrast, quinupristin binds in a similar manner to erythromycin at the proximal end of the tunnel, thereby accelerating the release of small oligopeptidyl-tRNAs [41]. When administered together, the two agents form a combination drug known as Synercid. It is used to treat staphylococcal infections [42].
Fig. 17.8

Streptogramins (A), daptomycin (B), and colistin (C)

Both type A and B streptogramins are susceptible to efflux-mediated resistance; indeed, the efflux protein Lsa intrinsic to Enterococcus faecalis confers resistance to type B streptogramins, limiting Synercid use [40]. A group of O-acetyltransferases confer high-level resistance to type A streptogramins, while Vgb is a ring-opening C–O lyase that provides resistance to type B antibiotics.


As their name suggests, these compounds are peptides that contain a lipid moiety. Both linear and cyclic, macrolactone and macrolactam lipopeptides exist. Due to the large variations in structure, these molecules have few well-characterized cellular targets. Daptomycin (Fig. 17.8b), initially discovered in the 1980s and discarded at Phase II clinical trials by Eli Lilly due to toxicity, was revisited with a new dosing regimen; it was approved for clinical use in 2003 [43]. Daptomycin has pleiotropic effects on the membrane of Gram-positive bacteria that result in depolarization and physical alteration of the cell membrane that leads to cell death [44]. Daptomycin is effective against most Gram-positive pathogens, including drug-resistant forms such as VRE and MRSA.

Colistin (polymyxin E) is a lipopeptide of the polymyxin class. It is one of the few antibiotics in clinical use that was derived from a non-actinomycete bacterium, Paenibacillus polymyxa. Discovered in 1949, colistin has been used sparingly for the treatment of serious infections caused by Gram-negative bacteria due to toxicity issues [45]. As a result of the rise of carbapenem-resistant Gram-negative pathogens, clinicians have been left with few therapeutic options other than colistin. Consequently, its use has increased significantly. The mode of action of polymyxins involves disruption of the outer and inner membranes of Gram-negative bacteria [46].

Resistance to colistin occurs through chemical modification of the lipopolysaccharide component of the Gram-negative outer membrane. Expression of intrinsic aminoarabinose transferases and phosphoethanolamine transferases, which modify lipid A impact the physical properties of the outer membrane, confers colistin resistance [47]. Mobilization of the mcr-1 phosphoethanolamine transferase gene in Gram-negative pathogens is a growing concern [48]. In contrast, daptomycin resistance remains rare and has not been mobilized. However, resistance mutations in cell membrane and cell-wall structure and function can be selected in vitro and during long-term therapy [49].


In 2007 the first pleuromutilin compound approved for clinical use was the topical antibiotic retapamulin [10]. Although retapamulin represents a new chemical scaffold for clinical use, it is actually a semisynthetic version of the original pleuromutilin, which was isolated from the fungus Pleurotus mutilis (renamed Clitopilus scyphoides in 1951) [50]. Pleuromutilins contain a fused 8-6-5 tricyclic diterpene architecture and an acyclic tail. It is among the few isoprenoid antibiotics to find clinical use (Fig. 17.9). These antibiotics bind to the peptidyl transfer center of the 50S ribosomal subunit, thereby blocking protein synthesis. 23S rRNA methyltransferases can confer resistance to this class of antibiotic. Retapamulin is a topical agent used for treatment of skin infections caused by Gram-positive bacteria; several other pleuromutilin derivatives are currently in various stages of clinical assessment for systemic use.
Fig. 17.9

Chloramphenicol, pleuromutilins, fosfomycin, and fidaxomicin

Although the isoprenoid class contains the most abundant natural products (over 50,000 known structures), very few are known to exhibit specific antibiotic activity. Other examples include platensimycin, platencin, and fusidic acid.


Chloramphenicol , discovered in 1947 in extracts of Streptomyces venezuelae, displays broad-spectrum bacteriostatic activity [51]. This small molecule contains a dichloroacetamide moiety and an aromatic nitro group (Fig. 17.9). The dichloroacetyl moiety is important for activity, as it impedes tRNA from binding to the peptidyl transferase in the 50S ribosomal subunit, thereby preventing elongation. Due to the low-cost production of chloramphenicol, this agent is frequently used in developing countries, even though it has been withdrawn from common use in many areas due to resistance and safety concerns, the latter resulting from low-frequency association with aplastic anemia. Resistance is primarily the result of O-acetyltransferases that modify the antibiotic.


The phosphonate fosfomycin is a small (138 Da) antibiotic produced by Streptomyces fradiae. It has been in clinical use since the early 1970s for the treatment of urinary tract infections. The rapid emergence of resistance has limited its use, but its high water solubility and low toxicity enables a single dose (3 g of drug/day) or very short course therapy. Fosfomycin targets bacterial cell-wall biosynthesis by inhibition of MurA, an enzyme involved in the first step in the biosynthesis of N-acetylmuramic acid [52].

Fosfomycin resistance is readily selected during therapy in the form of mutants in the glycerol-3-phosphate transporter, which is needed for fosfomycin entry into the cell but not essential for bacterial cell growth [52]. A series of enzymes, including glutathione transferases and epoxide hydrolases, are known to inactivate the antibiotic via opening the essential epoxide ring.


Fidaxomicin is the first member of the newest class of natural products to enter the market (2011). It consists of an 18-membered macrolactone polyketide that was discovered independently in Italy (lipiarrmycin), Japan (clostomicin), and the USA (tiacumicin) in the early 1970s [53]. The macrolactone is decorated with two acylated rhamnoses (Fig. 17.9). Fidaxomicin inhibits RNA polymerase by binding to a site distinct from the rifamycin binding site. There, fidaxomicin blocks the conversion of bound promoter DNA to the open single-strand complex that forms the transcription bubble. It has been approved for clinical use for treatment of Clostridium difficile-associated diarrhea.

17.3 Natural Products: A Privileged Source of Antibiotics

The vast majority of antibacterials used clinically are natural products, semisynthetic derivatives, or analogues thereof. As mentioned above, the only synthetic classes that are not derived from natural scaffolds are the sulfonamides, diaminopyrimidines, oxazolidinones, and quinolones. Natural products have always been a major source of human medicines and continue to be especially important as leads for antimicrobials – 74 of the 98 small molecules approved as antibacterials from 1981 to 2006 are natural products, semisynthetic derivatives, or natural-product mimics [10].

In the 1980s the arrival of combinatorial chemistry allowed the rapid synthesis of large numbers of synthetic compounds. That transformed the pharmaceutical industry, and companies began to favor screening vast libraries of synthetic compounds over natural-product extracts. With major advances in high-throughput screening (HTS) technologies through the 1990s and 2000s, companies obtained the ability to quickly screen libraries of millions of compounds. This strategy has been enormously successful in identifying lead compounds having targets in human cells, but it has not been successful for identifying new antibacterials. Extensive high-throughput screening campaigns at GlaxoSmithKline (GSK) [54] and AstraZeneca [55] both failed to identify any candidate structure worthy of further clinical development. The lack of success is due to a combination of several factors, with retrospective analyses of both campaigns pointing to a lack of chemical diversity in the compound libraries.

The chemical libraries of pharmaceutical companies have largely been constructed with guidance from Lipinski’s Rules [56], which aim to improve the likelihood of oral bioavailability by keeping molecular weights (MW) under 500, measures of hydrophobicity (logP or logD) less than 5, and the number of hydrogen-bond donors and acceptors in the molecule less than 10. However, antibiotics have long been known to occupy a different “chemical space” than other drugs, and often they exhibit multiple violations of Lipinski’s guidelines. An analysis of physiochemical properties by O’Shea and Moser in 2008 [57] compared a reference set of human drugs against compounds active against Gram-positive and Gram-negative bacteria. Average molecular weights were 338, 813, and 414, and average clogD7.4 values were 1.6, −0.2, and −2.8, respectively. Anti-Gram-positive compounds are more polar than reference drugs and can be much larger, especially if their targets are on the cell exterior (e.g., glycopeptides, lipopeptides). Compounds active against Gram-negative bacteria, which must cross the outer membrane, are much more polar and have a strict molecular weight cutoff at 600 Da, likely due to the limitations of transport through porins. Overington pointed out, however, that the bacterial target should be taken into account, since the physiochemical properties of antibiotics targeting the ribosome fall further outside Lipinski’s rules than antibacterials that have bacterial protein targets [58].

An analysis of 23 HTS campaigns at AstraZeneca, reported by Brown et al. [59], showed that active antibacterial project compounds were significantly more polar than the screening collection average. Improving biochemical potency through chemical modification of active leads often came with an increase in hydrophobicity and an increased probability of problematic plasma protein binding or cytotoxicity. In cases in which biochemical potency was maintained by increasing polarity, whole-cell activity remained elusive; designing polar compounds was not sufficient for antibacterial activity. Overall, the study highlights the complexities of bacterial cell penetration and efflux systems, especially in Gram-negative bacteria. The authors note that one possibility for improving the antibiotic chemical space of screening libraries would be to return to natural-product screening.

While there is little overlap in the chemical space of compounds in screening libraries with that of antibacterials, there is far more overlap in the physiochemical properties between antibiotics and natural products [60, 61]. In addition to hydrophilicity (i.e., log P), other properties, such as the number of rotatable bonds (molecular flexibility), polar surface area, H-bond donors and acceptors, molecular complexity, and 3-dimensionality [62, 63], are better represented in natural-product chemical space.

17.4 Traditional Natural-Product Discovery

Selman Waksman is credited for developing a procedure in which microbial exudates are screened for cell growth inhibition on the surface of solid agar medium plates. This method measures “zones of inhibition” around paper disks to which natural-product samples are applied [16]. The so-called Waksman Platform is much faster and more efficient than systematic testing for antibiotic efficacy in animal disease models, as performed by Ehrlich. When Waksman used the method for high-throughput analysis of soil microbe products, he discovered candicidin, the first polyene antifungal agent; streptomycin and neomycin, the first aminoglycosides; and many other agents that include streptothricin and actinomycin. Many clinically used antibiotics were subsequently found using this method: chlortetracycline (Lederle), chloramphenicol (Parke-Davis), erythromycin (Abbott and Lilly), and tetracycline (Pfizer) [64]. After successfully mining soil-derived bacteria, specifically streptomycetes, the returns dwindled as known compounds repeatedly surfaced in the screens [65]. Consequently, the natural-product screening programs of drug companies slowly shut down, and the focus switched to synthetic chemistry.

Traditionally, antibiotic discovery using the Waksman platform begins with a source of environmental microbes. These have been primarily obtained from soil samples collected by the employees, family, and associates of drug companies across the globe. The microbes in these samples generally focused on the actinomycetes, spore-forming bacteria that over the decades led to collections containing tens to hundreds of thousands of strains. The producer strains are typically grown in a variety of defined media, since the contents of the medium can significantly impact the production of a given compound. Following fermentation, organic solvent extracts or conditioned media samples are prepared and used for screening against a set of pathogenic bacteria. If a natural-product extract elicits antibiotic properties, then activity-guided purification is conducted to isolate the bioactive molecule, and the chemical structure of the active molecule is elucidated. If the chemical hit is promising, then semisynthetic or total synthetic variations of the lead compound are produced and tested. From hundreds of analogs, a therapeutic candidate may emerge. Large-scale production of the optimized lead compound is undertaken, and extensive safety tests are carried out before the candidate enters clinical trials. Three phases of appropriate clinical trials are performed, and if the candidate agent passes, it would proceed to the regulatory approval step. The discovery and development pipeline of an antibiotic can take upward of 10 years and cost hundreds of millions of dollars. Bérdy estimated that ~28,000 antimicrobial natural products from microbial sources have been reported using this approach. It is for this reason, along with the drought in discovery of antibiotics using other chemical matter, that many pharmaceutical companies have withdrawn investments and shut down antibiotic discovery programs [66].

17.5 The Future of Natural-Product Discovery

The time is right for a reevaluation of natural products in antibiotic drug discovery. Their historical success as drugs, the comparative shortcomings of screens of synthetic compound libraries, and the serious need for innovation in securing new antibiotics demand a fresh look at this source of bioactive chemistry. The rediscovery of well-known chemical scaffolds, which prompted a move away from natural products, can be mitigated in several ways. First, previously unsuccessful scaffolds can be reevaluated; second, successful antibiotic drugs can be reinvigorated by combining them with inhibitors of resistance and other antibiotic adjuvants; third, new scaffolds can be sourced from previously neglected genera or through mining of microbial genomes and metagenomes; and finally, “new-to-nature” compounds can be generated through synthetic biology strategies.

17.5.1 Revisiting Discarded Scaffolds

The three most recent natural-product antibiotics to enter the clinic, daptomycin, fidaxomicin, and the pleuromutilins, were all discovered and discarded decades before their successful clinical launch. In the case of daptomycin, off-target human toxicity was deemed a sufficient concern by Eli Lilly to halt clinical development. A decade later, with more careful drug dosing to avoid undesired effects, daptomycin was championed by Cubist Pharmaceuticals, which successfully brought the compound to market [43]. Fidaxomicin was discarded in the 1970s due to poor solubility and narrow spectrum, properties that are advantages in its new incarnation as an orally dosed drug to combat C. difficile [53]. These examples offer hope, perhaps even certainty, that new antibiotic drugs can be sourced from known compounds. The estimate that ~28,000 natural-product antibiotics have been reported, while fewer than 500 have entered into clinical use, is encouraging that we can revisit these compounds as antimicrobial sources.

There are challenges to this route, however. A practical consideration is that there is no ready way to obtain these compounds for reevaluation. Most were reported in the scientific or patent literature decades ago, but some remain in the yellowing lab books in the vaults of pharma. Unless the compounds progressed in the development process, the bacterial strains that produce them may not be available in public culture collections. The fate of the extensive libraries of producing organisms held by many companies active during the 1950s–1980s is not widely known. Some have been captured by new entities. For example, the historical Merck collections are now foundational resources of Fundación MEDINA and the Natural Products Discovery Institute. Most strain libraries , however, are not easy to access. This means that interesting chemical scaffolds may be lost until rediscovered by traditional screens. Another challenge is securing intellectual property on known natural compounds and their activities [67]. Nevertheless, a deep reservoir of knowledge and chemistry exists that can be tapped for twenty-first-century antibiotic drug discovery.

17.5.2 Natural-Product Adjuvants, Resistance Inhibitors, and Combination Therapies

All known antibiotic-producing microbes have multiple biosynthetic programs that generate additional natural products. In most cases, these appear to be unrelated to production of the antibiotic of interest, but there are examples in which the additional products are co-expressed to achieve improved antibiotic efficacy. Indeed, coproduction to achieve synergy between molecules may be commonplace in producing microbes [68]. This can include coproduction of nonantibiotic adjuvants that enhance antibiotic activity or inhibit resistance. It can also include co-expression of two (or more) antibiotic compounds that act synergistically. Examples of the latter are the streptogramin antibiotics (described above in Sect. 17.2). Streptogramin producers, such as Streptomyces pristinaespiralis, produce type A and B compounds in a ratio of ~7:3. Binding of the type A streptogramin to the peptidyl transferase center of the bacterial ribosome enhances binding of the type B antibiotic to the region of the peptide exit tunnel by ~100-fold, thereby accounting for the observed synergy (reviewed in Ref. 40).

Antibiotic adjuvants have little or no antimicrobial activity themselves, but they enhance antibiotic activity by facilitating transport or blocking resistance [69, 70]. The discovery of clavulanic acid, a potent inactivator of β-lactamases produced by the cephamycin C-producer Streptomyces clavuligerus, demonstrated that antibiotic producers can “protect their investments” by co-expressing inhibitors of resistance . Several other cephamycin producers also express clavulanic acid, suggesting that the strategy of producing both antibiotics and inhibitors of resistance may be common. We have prepared a cell-based platform for the screening of resistance inhibitors that can also be used in the rapid identification (and dereplication) of known antibiotic scaffolds [71]. Using this platform, we identified an inactivator of metallo-β-lactamases, including NDM-1 produced by a strain of Aspergillus versicolor [72]. This strain also has the biosynthetic machinery to produce a β-lactam antibiotic (unpublished observation). There is little doubt that many other antibiotic–adjuvant pairs exist in nature. Indeed, plant-derived natural products also show efficacy as adjuvants [73].

Screening for lethal synergy is another strategy for extending the life of antibiotics. For example, the combination of bacteriostatic inhibitors of gene expression , such as tetracycline, rifampicin, and chloramphenicol, with the bacteriostatic compound bicyclomycin (Fig. 17.10), an inhibitor of the Rho transcription terminator, caused rapid killing of Gram-negative bacteria [74]. Screens for antimicrobial synergy often use growth inhibition assays and select for increased bacteriostatic activity; however, time–kill assays can be employed to screen for lethal synergy combinations, which cause rapid killing and may diminish the rate of resistance.
Fig. 17.10

(A) Aspergillomarasmine A (AMA) is an inhibitor of metallo-β-lactamases (e.g., NDM-1) and is in preclinical development as an adjuvant with β-lactam antibiotics [72]. (B) Bicyclomycin, an inhibitor of Rho transcription terminator, exhibits lethal synergy with inhibitors of gene expression [74]. (C) Teixobactin is a cyclic depsipeptide that binds to lipid II [75]

The significant challenge in bringing such combinations to market is the need to match pharmacological and dosing properties for each component. This is not trivial and often cited as a complex barrier to systematic exploration of such pairs. Nevertheless, a combination strategy is routine in the treatment of cancer, HIV disease, and even bacterial infections such as tuberculosis.

17.5.3 Mining New Sources of Microbial Natural Products

The bulk of the actinomycetes screened by the pharmaceutical industry for antibiotic activities originate from soil environments. These sources are easy to access and offer a wide variety of conditions for enriching various genera. The question of whether such sampling reflects a reasonable representative distribution of microbial natural-product diversity is unresolved. The same common scaffolds can be readily found in samples from around the world, supporting the axiom that “everything is everywhere, the environment selects.” However, careful genomic sampling of signature natural-product biosynthetic elements, such as ketosynthase domains from polyketide synthases and adenylation domains of non-ribosomal peptide synthases from a variety of soil environments, suggests that indeed there are significant environmental differences in natural-product potential: there is significant chemical diversity still to be identified [76]. Indeed, marine actinomycetes are sources of several new compounds, many having biological activity (Fig. 17.11).
Fig. 17.11

Antibiotics produced by Gram-negative bacteria and marine bacteria

The fact that most of the focus in antibiotic natural-product discovery has been on the actinomycetes has prompted a search in other orders of bacteria. The Gram-negative betaproteobacteria, such as members of the genus Burkholderia, are prodigious producers of antibiotics [77]. The gamma-proteobacteria, pseudomonads, and the deltaproteobacteria, such as the Myxococci [78], also produce numerous natural-product antibiotics (Fig. 17.11). These have only just begun to be mined to discover new chemical scaffolds.

These sources though, are limited to strains that we can readily grow in the laboratory. The “great plate count anomaly” refers to the fact that we are generally limited to growing <5% of the detectable microbes in a soil sample. Strategies to mine this “microbial dark matter” offer ways to access new microbial genetic and chemical diversity [79]. An example of this approach is the iChip, a simple 96-compartment device to capture microbes and grow them in situ, with access to nutrients and growth factors of their natural environment [80]. Using this device and strategy, a new antibiotic, teixobactin, which represents a new scaffold, was identified from a previously uncultured bacterium [75]. Teixobactin, produced by a Gram-negative bacterium, has a Gram-positive-only profile. Its mode of action involves binding to lipid II, which is required for cell wall biosynthesis. Mining other difficult-to-grow bacteria for new chemistry should be possible and therefore offers hope that additional antibiotic scaffolds can be identified.

17.5.4 Genome and Metagenome Mining

Advancements in microbiology and molecular biology techniques have enabled the culture of microbes that previously were difficult to access. Advances in next-generation sequencing (NGS) are providing unequaled access to the genomic details of these organisms. Coupled with automated in silico prediction algorithms, such as antiSMASH, to identify biosynthetic gene clusters [81], this new genomic information has revealed a previously unappreciated and remarkable quantity and genetic diversity of natural products that can be (at least in principle) synthesized by microbes. On average, actinomycetes encode in their genomes 20–40 natural-product biosynthetic gene clusters; fungi encode even more. This new reality offers unprecedented opportunity to mine previously unknown or overlooked chemical diversity. Many of these compounds are difficult to detect and/or are found in low abundance. However, recent advances in mass spectrometry-based sampling and automated compound analysis and identification using artificial intelligence analysis (e.g., [82, 83, 84, 85]) enable rapid triage for novelty that was inaccessible from traditional activity-guided purification and characterization methods. Such approaches are yielding new antibiotic scaffolds, such as the telomycins (Fig. 17.12), that target components of the bacterial membrane [86].
Fig. 17.12

Telomycin and turbomycins A and B [86, 87]

Often the expression of biosynthetic gene clusters in the laboratory is challenging, thereby preventing testing or purification of new compounds. Strategies to activate such “silent” clusters are being explored, although none is universal [88, 89, 90, 91]. This includes the deletion or overexpression of regulatory genes, addition of chemical perturbants, physical stress (e.g., pH, temperature), and selection of mutants of various genes, such as encoded ribosomal proteins and antibiotic resistance. Failing such strategies, capture of entire clusters and mobilization to surrogate hosts can be used [92, 93].

While genome mining has greatly expanded our access to known and new antibiotic scaffolds, the majority of environmental microbes remain difficult to culture. Here, metagenomic strategies in which total DNA is collected from a source (e.g., a soil sample, animal, or plant microbiomes) are being employed. Such strategies are yielding new antimicrobial compounds, such as the turbomycins [87], variants of glycopeptides [94], and colicins [95] (Fig.17.12).

17.5.5 Increasing Diversity Through Synthetic Biology

The long-term future for obtaining antibiotic diversity may be through the generation of nonnatural or synthetic natural products . This oxymoron refers to the engineering of biosynthetic gene clusters to produce novel compounds, not yet known to nature, using synthetic biology concepts [96, 97, 98]. The modularity of biosynthetic gene clusters lends itself well to systematic synthetic biology. Biosynthetic gene clusters include a predictable parts list: genes encoding scaffold assembly, tailoring enzymes, supply of components not easily scavenged from primary metabolism (amino acids, sugars, etc.), self-resistance, regulation, and transport (Fig. 17.13). In principle, these elements can be mixed and matched to generate new compounds having novel activities. For example, we have used this approach to generate new-to-nature glycopeptide antibiotics that evade certain forms of resistance in VRE [99] (Fig. 17.13).
Fig. 17.13

A synthetic biology approach for increasing chemical diversity in glycopeptide antibiotics. (A) Biosynthetic clusters for selected glycopeptides with genes colored according to function. Significant portions of many clusters are comprised of tailoring genes responsible for decorating the glycopeptide backbone (e.g., methyltransferases, sulfotransferases, and glycosyltransferases). (B) Novel glycopeptides have been generated by mixing biosynthetic genes. In a recent study by Yim et al. [99], the biosynthetic clusters for A47934 and desulfo-A47934 were expressed in S. coelicolor together with a variety of tailoring genes from the biosynthetic clusters of other glycopeptides. Several new A47934- and desulfo-A47934 derivatives were produced and are more potent than vancomycin against E. faecalis and VRE B (C)

As the costs of DNA synthesis continue to drop , one can envision synthesis of large numbers of biosynthetic gene cluster parts , the combinatorial generation of libraries of scaffolds, tailoring enzymes, regulatory elements, etc., and their expression in a suitable heterologous host. The result would be millions of previously untested combinations of biosynthetic genes (Fig. 17.13). With suitable selection, such libraries could deliver hits and lead for new antibiotic drug development.

17.6 Concluding Remarks

Natural products, in particular those generated by bacteria and fungi, are the source of the majority of our successful antibiotic drugs. These agents have changed the world of medicine. For the first time in human history, we have good control over infection. With that control has come much of modern medicine. Our natural-product antibiotics have also helped us feed the world by changing the way we raise and care for food animals. It is not hyperbole to suggest that natural-product antibiotics may be the most important scientific discovery of the twentieth century.

Unfortunately, the evolution of antibiotic resistance and its selection in once-susceptible pathogens gravely threatens these advances. We need new antibiotics and alternatives to maintain our control over infectious disease. The advances in our knowledge of how natural products are made by microbes, new and unparalleled access to the genetic determinants of natural-product biosynthesis by NGS of microbial genomes and metagenomes, and the ability to harness this information to identify and exploit this information are growing exponentially.

The proven efficacy of natural products as antibiotics, plus the disappointing results of the past two decades of focus on synthetic compounds, means that we must pivot back to these ancient compounds for leads and inspiration. There is good reason to believe that the era of resistance depicted in Fig. 17.1 will be followed by an era of anti-infective innovation.

Major Points

  • Microbial natural products are the source of most of our successful antibiotic drugs.

  • These compounds are the result of evolutionary processes that select for optimal penetration and retention in target bacterial cells.

  • The chemical diversity and physiochemical properties of microbial natural products cannot yet be effectively matched in most synthetic libraries.

  • The re-isolation of known natural-product scaffolds diminished enthusiasm for the natural-product approach in antibiotic discovery.

  • Efforts to identify new antibiotic chemical diversity through revisiting discarded compounds, mining of bacterial genomes, isolation of hitherto rare or unsampled microbes, and increasing chemical diversity using synthetic biology strategies offer new routes to identifying antibiotic leads.

  • The use of inhibitors of resistance or other adjuvants can also extend the clinical effectiveness of existing antibiotic scaffolds.


  1. 1.

    Here we are deviating from Waksman’s definition of antibiotics restricted narrowly to compounds synthesized by microbes to include synthetic and semisynthetic human-made compounds as well.


  1. 1.
    Borchardt JK. The beginnings of drug therapy: ancient Mesopotamian medicine. Drug News Perspect. 2002;15(3):187–92.PubMedGoogle Scholar
  2. 2.
    Scurlock J. Sourcebook for ancient Mesopotamian medicine: Society of Biblical Literature; 2014.Google Scholar
  3. 3.
    Cragg GM, Newman DJ. Natural products: a continuing source of novel drug leads. Biochim Biophys Acta. 2013;1830(6):3670–95.PubMedPubMedCentralGoogle Scholar
  4. 4.
    Huang KC. The pharmacology of chinese herbs. 2nd ed. Boca Raton: CRC Press; 1999.Google Scholar
  5. 5.
    Kapoor LD. CRC handbook of ayurvedic medicinal plants. Boca Raton: CRC Press; 1990.Google Scholar
  6. 6.
    Sertuerner F. Ueber das Morphium, eine neue salzfähige Grundlage, und die Mekonsäure, als Hauptbestandtheile des Opiums. Ann Phys (Berl). 1817;55(1):56–89.Google Scholar
  7. 7.
    Newman DJ, Cragg GM. Chapter 1 Natural products as drugs and leads to drugs: the historical perspective. Natural Product Chemistry for Drug Discovery: The Royal Society of Chemistry; 2009. p. 3–27.Google Scholar
  8. 8.
    Fleming A. On the antibacterial action of cultures of a Penicillium, with special reference to their use in the isolation of B. influenzæ. Br J Exp Pathol. 1929;10(3):226–36.PubMedCentralGoogle Scholar
  9. 9.
    Zasloff M. Antimicrobial peptides of multicellular organisms. Nature. 2002;415(6870):389–95.PubMedGoogle Scholar
  10. 10.
    Newman DJ, Cragg GM. Natural products as sources of new drugs from 1981 to 2014. J Nat Prod. 2016;79(3):629–61.PubMedGoogle Scholar
  11. 11.
    Walsh CT, Wencewicz T. Antibiotics: challenges mechanisms opportunities. Washington, DC: ASM Press; 2016.Google Scholar
  12. 12.
    Brown ED, Wright GD. Antibacterial drug discovery in the resistance era. Nature. 2016;529(7586):336–43.PubMedGoogle Scholar
  13. 13.
    Ehrlich P, Hata S. Die experimentelle Chemotherapie der Spirillosen:(Syphilis, Rückfallfieber, Hühnerspirillose, Frambösie). Wiesbaden: Springer; 1910.Google Scholar
  14. 14.
    Aminov RI. A brief history of the antibiotic era: lessons learned and challenges for the future. Front Microbiol. 2010;1:134.PubMedPubMedCentralGoogle Scholar
  15. 15.
    Mahoney JF, Arnold RC, Harris A. Penicillin treatment of early syphilis-a preliminary report. Am J Public Health Nations Health. 1943;33(12):1387–91.PubMedPubMedCentralGoogle Scholar
  16. 16.
    Lewis K. Platforms for antibiotic discovery. Nat Rev Drug Discov. 2013;12(5):371–87.PubMedGoogle Scholar
  17. 17.
    Domagk G. Ein Beitrag zur Chemotherapie der bakteriellen Infektionen. Dtsch Med Wochenschr. 1935;61(7):250–3.Google Scholar
  18. 18.
    Lewis K. Antibiotics: recover the lost art of drug discovery. Nature. 2012;485(7399):439–40.PubMedGoogle Scholar
  19. 19.
    Wright GD. Opportunities for natural products in 21st century antibiotic discovery. Nat Prod Rep. 2017;34(7):694–701.PubMedGoogle Scholar
  20. 20.
    Tipper DJ, Strominger JL. Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl Acad Sci U S A. 1965;54(4):1133–41.PubMedPubMedCentralGoogle Scholar
  21. 21.
    Rolinson GD. Forty years of beta-lactam research. J Antimicrob Chemother. 1998;41(6):589–603.PubMedGoogle Scholar
  22. 22.
    Drawz SM, Papp-Wallace KM, Bonomo RA. New β-lactamase inhibitors: a therapeutic renaissance in an MDR world. Antimicrob Agents Chemother. 2014;58(4):1835–46.PubMedPubMedCentralGoogle Scholar
  23. 23.
    Wang DY, Abboud MI, Markoulides MS, Brem J, Schofield CJ. The road to avibactam: the first clinically useful non-β-lactam working somewhat like a β-lactam. Future Med Chem. 2016;8(10):1063–84.PubMedGoogle Scholar
  24. 24.
    Castanheira M, Rhomberg PR, Flamm RK, Jones RN. Effect of the β-lactamase inhibitor Vaborbactam combined with Meropenem against serine carbapenemase-producing Enterbacteriaceae. Antimicrob Agents Chemother. 2016;60(9):5454–8.PubMedPubMedCentralGoogle Scholar
  25. 25.
    Sykes RB, Cimarusti CM, Bonner DP, Bush K, Floyd DM, Georgopapadakou NH, Koster WM, Liu WC, Parker WL, Principe PA, Rathnum ML, Slusarchyk WA, Trejo WH, Wells JS. Monocyclic beta-lactam antibiotics produced by bacteria. Nature. 1981;291(5815):489–91.PubMedGoogle Scholar
  26. 26.
    Sykes RB, Bonner DP, Bush K, Georgopapadakou NH. Azthreonam (SQ26,776), a synthetic monobactam specifically active against aerobic gram-negative bacteria. Antimicrob Agents Chemother. 1982;21(1):85–92.PubMedPubMedCentralGoogle Scholar
  27. 27.
    Waksman SA, Reilly HC, Johnston DB. Isolation of streptomycin-producing strains of Streptomyces griseus. J Bacteriol. 1946;52(3):393–7.PubMedPubMedCentralGoogle Scholar
  28. 28.
    Umezawa H. Kanamycin: its discovery. Ann N Y Acad Sci. 1958;76(2):20–6.PubMedGoogle Scholar
  29. 29.
    Weinstein MJ, Luedemann GM, Oden EM, Wagman GH, Rosselet JA, Coniglio CT, Charney W, Herzog HL, Black J. Gentamicin, a new antibiotic complex from micromonospora. J Med Chem. 1963;6(4):463.PubMedGoogle Scholar
  30. 30.
    Garneau-Tsodikova S, Labby KJ. Mechanisms of resistance to aminoglycoside antibiotics: overview and perspectives. Medchemcomm. 2016;7(1):11–27.PubMedGoogle Scholar
  31. 31.
    Katz L, Ashley GW. Translation and protein synthesis: macrolides. Chem Rev. 2005;105(2):499–528.PubMedGoogle Scholar
  32. 32.
    Gomes C, Martinez-Puchol S, Palma N, Horna G, Ruiz-Roldán L, Pons MJ, Ruiz J. Macrolide resistance mechanisms in Enterobacteriaceae: focus on azithromycin. Crit Rev Microbiol. 2017;43(1):1–30.PubMedGoogle Scholar
  33. 33.
    Wright PM, Seiple IB, Myers AG. The evolving role of chemical synthesis in antibacterial drug discovery. Angew Chem Int Ed Engl. 2014;53(34):8840–69.PubMedPubMedCentralGoogle Scholar
  34. 34.
    Yang W, Moore IF, Koteva KP, Bareich DC, Hughes DW, Wright GD. TetX is a flavin-dependent monooxygenase conferring resistance to tetracycline antibiotics. J Biol Chem. 2004;279(50):52346–52.PubMedGoogle Scholar
  35. 35.
    Floss HG, Yu TW. Rifamycin-mode of action, resistance, and biosynthesis. Chem Rev. 2005;105(2):621–32.PubMedGoogle Scholar
  36. 36.
    Goldstein BP. Resistance to rifampicin: a review. J Antibiot (Tokyo). 2014;67(9):625–30.Google Scholar
  37. 37.
    Bugg TD, Wright GD, Dutka-Malen, S, Arthur, M, Courvalin, P, Walsh CT. Molecular basis for vancomycin resistance in Enterococcus faecium BM4147: biosynthesis of a depsipeptide peptidoglycan precursor by vancomycin resistance proteins VanH and VanA. Biochem. 1991; 30(43):10408–10415.Google Scholar
  38. 38.
    Kirst HA, Thompson DG, Nicas TI. Historical yearly usage of vancomycin. Antimicrob Agents Chemother. 1998;42(5):1303–4.PubMedPubMedCentralGoogle Scholar
  39. 39.
    Kahne D, Leimkuhler C, Lu W, Walsh C. Glycopeptide and lipoglycopeptide antibiotics. Chem Rev. 2005;105(2):425–48.PubMedGoogle Scholar
  40. 40.
    Mukhtar TA, Wright GD. Streptogramins, oxazolidnones, and other inhibitors of bacgerial protein synthesis. Chem Rev. 2005;105(2):529–42.PubMedGoogle Scholar
  41. 41.
    Wilson DN. The A-Z of bacterial translation inhibitors. Crit Rev Biochem Mol Biol. 2009;44(6):393–433.PubMedGoogle Scholar
  42. 42.
    Allington DR, Rivey MP. Quinupristin/dalfopristin: a therapeutic review. Clin Ther. 2001;23(1):24–44.PubMedGoogle Scholar
  43. 43.
    Eisenstein BI, Oleson FB Jr, Baltz RH. Daptomycin: from the mountain to the clinic, with essential help from Francis Tally, MD. Clin Infect Dis. 2010;50(1):S10–5.PubMedGoogle Scholar
  44. 44.
    Taylor SD, Palmer M. The action mechanism of daptomycin. Bioorg Med Chem. 2016;24(24):6253–68.PubMedGoogle Scholar
  45. 45.
    Ordooei Javan A, Shokouhi S, Sahraei Z. A review on colisten nephrotoxicity. Eur J Clin Pharmacol. 2015;71(7):801–10.PubMedGoogle Scholar
  46. 46.
    Velkov T, Thompson PE, Nation RL, Li J. Structure-activity relationships of polymyxin antibiotics. J Med Chem. 2010;53(5):1898–916.PubMedPubMedCentralGoogle Scholar
  47. 47.
    Yu Z, Quin W, Lin J, Fang S, Qiu J. Antibacterial mechanisms of polymyxin and bacterial resistance. Biomed Res Int. 2015;2015:679109.PubMedPubMedCentralGoogle Scholar
  48. 48.
    Srinivas P, Rivard K. Polymyxin resistance in gram-negative pathogens. Curr Infect Dis Rep. 2017;19(11):38.PubMedGoogle Scholar
  49. 49.
    Miller WR, Bayer AS, Arias CA. Mechanism of action and resistance to daptomycin in Staphylococcus aureus and Enterocci. Cold Spring Harb Perspect Med. 2016;6(11):a026997.PubMedPubMedCentralGoogle Scholar
  50. 50.
    Novak R. Are pleuromutilin antibiotics finally fit for human use? Ann N Y Acad Sci. 2011;1241:71–81.PubMedGoogle Scholar
  51. 51.
    Dinos GP, Athanassopoulos CM, Missiri DA, Giannopoulou PC, Vlachogiannis IA, Papadopoulos GE, Papaioannou D, Kalpaxis DL. Chloramphenicol derivatives as antibacterial and anticancer agents: historic problems and current solutions. Antibiotics (Basel). 2016;5(2):20.Google Scholar
  52. 52.
    Silver LL. Fosfomycin: mechanism and resistance. Cold Spring Harb Perpect Med. 2017;7(2):a025262.Google Scholar
  53. 53.
    Fidaxomicin. A novel agent for the treatment of Clostridium difficile infection. Can J Infect Dis Med Microbiol. 2015;26(6):305–12.Google Scholar
  54. 54.
    Payne DJ, Gwynn MN, Holmes DJ, Pompliano DL. Drugs for bad bugs: confronting the challenges of antibacterial discovery. Nat Rev Drug Discov. 2007;6(1):29–40.PubMedGoogle Scholar
  55. 55.
    Tommasi R, Brown DG, Walkup GK, Manchester JI, Miller AA. ESKAPEing the labyrinth of antibacterial discovery. Nat Rev Drug Discov. 2015;14(8):529–42.PubMedGoogle Scholar
  56. 56.
    Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rec. 2001;46(1–3):3–26.Google Scholar
  57. 57.
    O'Shea R, Moser HE. Physicochemical properties of antibacterial compounds: implications for drug discovery. J Med Chem. 2008;51(10):2871–8.PubMedGoogle Scholar
  58. 58.
    Mugumbate G, Overington JP. The relationship between target-class and the physiochemical properties of antibacterial drugs. Bioorg Med Chem. 2015;23(16):5218–24.PubMedPubMedCentralGoogle Scholar
  59. 59.
    Brown DG, Tl M-D, Gagnon MM, Tommasi R. Trends and exceptions of physical properties on antibacterial activity for Gram-positive and Gram-negative pathogens. J Med Chem. 2014;57(23):10144–61.PubMedGoogle Scholar
  60. 60.
    Ganesan A. The impacet of natural product upon modern drug discovery. Curr Opin Chem Biol. 2008;12(3):306–17.PubMedGoogle Scholar
  61. 61.
    Harvey AL, Edrada-Ebel R, Quinn RJ. The re-emergene of natural products for drug discovery in the genomics era. Nat Rev Drug Discov. 2015;14(2):111–29.PubMedGoogle Scholar
  62. 62.
    Lovering F, Bikker J, Humblet C. Escape from flatland: increasing saturation as an approach to improving clinial success. J Med Chem. 2009;52(21):6752–6.PubMedGoogle Scholar
  63. 63.
    Lovering F. Escape from Flatland 2: complexity and promiscuity. Med Chem Commun. 2013;4(3):515–9.Google Scholar
  64. 64.
    Buggs CW. Ten years after streptomycin; past and current practice in antibiotic therapy. J Natl Med Assoc. 1957;49(3):142–9.PubMedPubMedCentralGoogle Scholar
  65. 65.
    Baltz RH. Antimicrobials from actinomycetes: back to the future. Microbe. 2007;2(3):125.Google Scholar
  66. 66.
    Katz L, Baltz RH. Natural product discovery: past, present, and future. J Ind Microbiol Biotechnol. 2016;43(2–3):155–76.PubMedGoogle Scholar
  67. 67.
    Chang B-C, Wang S-J. The impact of patent eligibility on biotech patents: a flow chart for determining patent eligibility and an immune therapy case study. Hum Vaccin Immunother. 2015;11(3):789–94.PubMedPubMedCentralGoogle Scholar
  68. 68.
    Challis GL, Hopwood DA. Synergy and contingency as driving forces for the evolution of multiple secondary metabolite production by Streptomyces species. Proc Natl Acad Sci U S A. 2003;100(2):14555–61.PubMedPubMedCentralGoogle Scholar
  69. 69.
    Wright GD. Antibiotic adjuvants: rescuing antibiotics from resistance. Trends Microbiol. 2016;24(11):862–71.PubMedGoogle Scholar
  70. 70.
    Melander RJ, Melander C. The challenge of overcoming antibiotic resistance: an adjuvant approach? ACS Infect Dis. 2017;3(8):559–63.PubMedPubMedCentralGoogle Scholar
  71. 71.
    Cox G, Siron A, King AM, De Pascale G, Pawlowski AC, Koteva K, Wright GD. Cell Chem Biol. 2017;24(1):98–109.PubMedGoogle Scholar
  72. 72.
    King AM, Reid-Yu SA, Wang W, King DT, De Pascale G, Strynadka NC, Walsh TR, Coombes BK, Wright GD. Aspergillomarasmine A overcomes metallo-β-lacamse antibiotic resistance. Nature. 2014;510(7506):503–6.PubMedPubMedCentralGoogle Scholar
  73. 73.
    Stermitz FR, Lorenz P, Tawara JN, Zenewicz LA, Lewis K. Synergy in a medicinal plant: antimicrobial action of berberine potentiated by 5′-methoxyhydnocarpin, a multidrug pump inhibitor. Proc Natl Acad Sci U S A. 2000;97(4):1433–7.PubMedPubMedCentralGoogle Scholar
  74. 74.
    Malik M, Li L, Zhao X, Kerns RJ, Berger JM, Drlica K. Lethal synergy involving bicyclomycin: an approach for reviving old antibiotics. J Antimicrob Chemother. 2014;69(12):3227–35.PubMedPubMedCentralGoogle Scholar
  75. 75.
    Ling LL, Schneider T, Peoples AJ, Spoering AL, Engels I, Conlon BP, Mueller A, Schäberle TF, Hughes DE, Epstein S, Jones M, Lazarides L, Steadman VA, Cohen DR, Felix CR, Fetterman KA, Millet WP, Nitti AG, Zullo AM, Chen X, Lewis K. A new antibiotic kills pathogens without detectable resistance. Nature. 2015;517(7535):455–9.PubMedGoogle Scholar
  76. 76.
    Charlop-Powers Z, Owen JG, Reddy BV, Ternei MA, Guimarães DO, de Frias UA, Pupo MT, Seepe P, Feng Z, Brady SF. Global biogeographical sampling of bacterial secondary metabolism. elife. 2015;4:e05048.PubMedPubMedCentralGoogle Scholar
  77. 77.
    Masschelein J, Jenner M, Challis GL. Antibiotics from Gram-negative bacteria: a comprehensive overview and selected biosynthetic highlights. Nat Prod Rep. 2017;34(7):712–83.PubMedGoogle Scholar
  78. 78.
    Hermann J, Fayad AA, Müller R. Natural products from myxobacteria: novel metabolites and bioactivities. Nat Prod Rep. 2017;34(2):135–60.Google Scholar
  79. 79.
    Lok C. Mining the microbial dark matter. Nature. 2015;522(7556):270–3.PubMedGoogle Scholar
  80. 80.
    Nichols D, Cahoon N, Trakhtenber EM, Pham L, Mehta A, Belanger A, Kanigan T, Lewis K, Epstein SS. Use of ichip for high-throughput in situ cultivation of “uncultivable” microbial species. Appl Environ Microbiol. 2010;76(8):2445–50.PubMedPubMedCentralGoogle Scholar
  81. 81.
    Blin K, Wolf T, Chevrette MG, Lu X, Schwalen CJ, Kautsar SA, Suarez Duran HG, de Los Santos ELC, Kim H, Nave M, Dickschat JS, Mitchell DA, Shelest E, Breitling R, Takano E, Sy L, Webe T, Medema MH. antiSMASH 4.0-improvements in chemistry predictions and gene cluster boundary identification. Nucleic Acids Res. 2017;45(W1):W36–41. Scholar
  82. 82.
    Ibrahim A, Yang L, Johnston C, Liu X, Ma B, Magarvey NA. Dereplicating nonribosomal peptides using an informatic search algorithm for natural products (iSNAP) discovery. Proc Natl Acad Sci U S A. 2012;109(47):19196–201.PubMedPubMedCentralGoogle Scholar
  83. 83.
    Johnston CW, Skinnider MA, Wyatt MA, Li X, Ranieri MR, Yang L, Zecehl DL, Ma B, Magarvey NA. An automated Gencomes-to-Natural Products platform (GNP) for the discovery of modular natural products. Nat Commun. 2015;8:8421.Google Scholar
  84. 84.
    Medema MH, Osbourn A. Computational genomic identification and functional reconstitution of plant natural product biosynthetic pathways. Nat Prod Rep. 2016;33(8):951–62.PubMedPubMedCentralGoogle Scholar
  85. 85.
    Skinnifer MA, Merwin NJ, Johnston CW, Magarvey NA. PRISM 3: expanded prediction of natural product chemical structures from microbial genomes. Nucleic Acids Res. 2017;45(W1):W49–54.Google Scholar
  86. 86.
    Johnston CW, Skinnider MA, Ca D, Rees PN, Chen GM, Walker CG, French S, Brown ED, Bérdy J, Liu DY, Magarvey NA. Assembly and clustering of natural antibiotics guides target identification. Nat Chem Biol. 2016;12(4):233–9.PubMedGoogle Scholar
  87. 87.
    Gillespie DE, Brady SF, Bettermann AD, Cianciotto NP, Liles MR, Rondon MR, Clardy J, Goodman RM, Handelsman J. Isolation of antibiotics turbomycin A and B from a metagenomic library of soil microbial DNA. Appl Environ Microbiol. 2002;68(9):4301–6.PubMedPubMedCentralGoogle Scholar
  88. 88.
    Tanaka Y, Tokuyama S, Ochi K. Activation of secondary metabolite-biosynthetic gene clusters by generating rsmG mutations in Streptomyces griseus. J Antibiot (Tokyo). 2009;62(12):669–73.Google Scholar
  89. 89.
    Tanaka Y, Kasahara K, Hirose Y, Murakami K, Kugimiya R, Ochi K. Activation and products of the cryptic secondary metabolite biosynthetic gene clusters by rifampin resistance (rpoB) mutations in actinomycetes. J Bacteriol. 2012;195(10):2959–70.Google Scholar
  90. 90.
    Yoon V, Nodwell JR. Activating secondary metabolism with stress and chemicals. J Ind Microbiol Biotechnol. 2014;41(2):415–24.PubMedGoogle Scholar
  91. 91.
    Onaka H. Novel antibiotic screening methods to awaken silent or cryptic secondary metabolic pathways in actinomycetes. J Antibiot (Tokyo). 2017;70(8):865–70.Google Scholar
  92. 92.
    Yamanaka K, Reynolds KA, Kersten RD, Tyan KS, Gonzalez DJ, Nizet V, Dorrestein PC, Moore BS. Direct cloning and refactoring of a silent lipopeptide biosynthetic gene cluster yields the antibiotic taromycin A. Proc Natl Acad Sci U S A. 2014;111(5):1957–62.PubMedPubMedCentralGoogle Scholar
  93. 93.
    Li Y, Li Z, Yamanaka K, Xy Y, Zhang W, Wlamakis H, Kolter R, Moore BS, Qian PY. Directed natural product biosynthesis gene cluster capture and expression in the model bacterium Bacilus subtilis. Sci Rep. 2015;5(9383). Google Scholar
  94. 94.
    Banik JJ, Brady SF. Cloning and characterization of new glycopeptide gene clusters found in environmental DNA megalibrary. Proc Natl Acad Sci U S A. 2008;105(45):17273–7.PubMedPubMedCentralGoogle Scholar
  95. 95.
    Cohen LJ, Han S, Huang YH, Brady SF. Identification of the colicin V bacteriocin gene cluster by functional screening of a human microbiome metagenomic library. ACS Infec Dis; 2017. 4(1):27–32.Google Scholar
  96. 96.
    Wright GD. Perspective: synthetic biology revives antibiotics. Nature. 2014;509(7498):S13.PubMedGoogle Scholar
  97. 97.
    Thaker MN, Wright GD. Opportunities for synthetic biology in antibiotics:expanding glycopeptide diversity. ACS Synth Biol. 2015;4(3):195–206.PubMedGoogle Scholar
  98. 98.
    Braff D, Shis D, Collins JJ. Synthetic biology platform technologies for antimicrobial applications. Adv Drug Deliv Rev. 2016;105(Pt A):35–43.PubMedGoogle Scholar
  99. 99.
    Yim G, Wang W, Thaker MN, Tan S, Wriight GD. How to make a glycopeptide: asynthetic biology approach to expand glycopeptide antibiotic chemcial diversity. ACS Infect. Dis. 2016;2(9):642–50.PubMedGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Fern R. McSorley
    • 1
  • Jarrod W. Johnson
    • 1
  • Gerard D. Wright
    • 1
    Email author
  1. 1.Department of Biochemistry and Biomedical SciencesMcMaster University, Michael G. DeGroote Institute for Infectious Disease Research, McMaster UniversityHamiltonCanada

Personalised recommendations