Review
Nucleoside antibiotics: biosynthesis, regulation, and biotechnology Guoqing Niu and Huarong Tan State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, No 1 Beichen West Road, Chaoyang District, Beijing 100101, China
The alarming rise in antibiotic-resistant pathogens has coincided with a decline in the supply of new antibiotics. It is therefore of great importance to find and create new antibiotics. Nucleoside antibiotics are a large family of natural products with diverse biological functions. Their biosynthesis is a complex process through multistep enzymatic reactions and is subject to hierarchical regulation. Genetic and biochemical studies of the biosynthetic machinery have provided the basis for pathway engineering and combinatorial biosynthesis to create new or hybrid nucleoside antibiotics. Dissection of regulatory mechanisms is leading to strategies to increase the titer of bioactive nucleoside antibiotics. Nucleoside antibiotics Nucleoside antibiotics are a large family of microbial natural products derived from nucleosides and nucleotides. Because nucleosides and nucleotides play essential roles in most of the fundamental cellular metabolism, nucleoside antibiotics exhibit a broad spectrum of biological activities, such as antibacterial, antifungal, antiviral, insecticidal, immunostimulative, immunosuppressive, and antitumor activities [1]. Based on their biological functions, they can be classified into three major groups. The antibacterial nucleoside antibiotics target bacterial cell wall biosynthesis: they are competitive inhibitors of bacterial phospho-N-acetylmuramyl-pentapeptide translocase (translocase I, denoted MraY), which initiates the lipid cycle of peptidoglycan biosynthesis [2]. The antifungal nucleoside antibiotics also target cell wall biosynthesis: they act as competitive inhibitors of fungal chitin synthases. They can also function as inhibitors of protein biosynthesis. The antiviral nucleoside antibiotics block protein biosynthesis by inhibiting peptidyl transferase. Over the past two decades, several excellent reviews have focused on the structure, biosynthesis, and biological activity of nucleoside antibiotics [1–3]. Here we highlight recent findings on their biosynthesis and its complex regulation and summarize progress in the use of this information to generate new nucleoside antibiotics and to increase production levels.
Corresponding author: Tan, H. ([email protected]). Keywords: nucleoside antibiotics; biosynthesis; regulation; genetic engineering. 0966-842X/ ß 2014 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.tim.2014.10.007
110
Trends in Microbiology, February 2015, Vol. 23, No. 2
Biosynthesis of nucleoside antibiotics Nucleoside antibiotics are endowed with unique structural features, suggesting the occurrence of unusual enzymatic reactions during the biosynthetic process. With the rapid development of methodology and the availability of an increasing number of gene clusters, biosynthetic studies have accelerated at both genetic and biochemical levels, providing the basis for pathway engineering and combinatorial biosynthesis of nucleoside antibiotics. Biosynthesis of antibacterial nucleoside antibiotics: pacidamycin, liposidomycin, tunicamycin, and capuramycin This group, which is also referred to as uridine-based nucleosides, can be further divided into four subfamilies: uridyl peptide antibiotics, uridyl lipopeptide antibiotics, uridyl lipodisaccharide antibiotics, and uridyl glycosylpeptide antibiotics. The uridyl peptide antibiotics encompass pacidamycin, napsamycin, mureidomycin, and sansanmycin. They share a common structural scaffold, a unique 30 -deoxy40 ,50 -enamino-uridine nucleoside linked to a pseudotetraor pentapeptide backbone (Figure 1), and exhibit selective antibacterial activity against Pseudomonas aeruginosa, a common nosocomial pathogen that is intrinsically resistant to various clinically used antibiotics. Sansanmycin also displays inhibition against multidrug-resistant Mycobacterium tuberculosis strains [4]. The uridyl lipopeptide antibiotics are represented by liposidomycin, caprazamycin, muraymycin, muraminomicin, and A-90289. They are structurally characterized by a 50 -C-glycyluridine (GlyU), an aminoribofuranoside, a diazepanone, and variable fatty acid side chains (Figure 1). The uridyl lipodisaccharide antibiotics are represented by tunicamycins comprising an unusual 11-carbon aminodialdose core, uracil, N-acetylglucosamine (GlcNAc), and variable fatty acyl moieties (Figure 1). The uridyl glycosylpeptide antibiotics include capuramycin and related compounds (A-503083, A-102395, and A-500359), which are characterized by a 50 -carbamoyluridine (CarU), an unsaturated hexuronic acid, and an aminocaprolactam ring (Figure 1). Comparison of gene clusters shows high sequence similarities of the corresponding proteins involved in the biosynthesis of pacidamycin, napsamycin/mureidomycin, and sansanmycin [5–8]. Most genetic and biochemical evidence comes from pacidamycin biosynthesis. Biosynthesis of the peptide backbone has been investigated extensively by Christopher Walsh and colleagues. Apart from the two non-proteinogenic amino acids meta-tyrosine (m-Tyr)
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
R1
AA 2
O
N
AA 5
H N
AA4
O
O N-Methyl-DABA
TIC2
m-Tyr Ala m-Tyr – m-Tyr Gly Ala/TIC 1 – TIC 2/TIC 3 – TIC 2/TIC 3 – –/Gly m-Tyr –/Gly m-Tyr m-Tyr – m-Tyr – TIC 4 – m-Tyr –
Pacidamycin 1/2/3: Pacidamycin 4/5/5T: Pacidamycin 6/7: Pacidamycin D/4N: Napsamycin A/B: Napsamycin C/D: MureidomycinA/C: MureidomycinB/D: Sansanmycin A/B/C: Sansanmycin D/E: Sansanmycin F/G: Sansanmycin H:
NH O
N
RO
O
OH
OH
AA 2
L1
n
O
NH
NH
R1
O
N
O
N
N H
O
O
X
H N
OH
H O
O
OH
TIC 1
O
H N
O
O
NH
NH NH
X
O
O
OH O
N
HN HO
TIC 4 HO
AA 4
AA 5
X-X
Ala Ala Ala Ala Met Met Met Met Met/Leu/MetSO Met/MetSO Met/Leu Met
Trp/Phe/ m-Tyr Trp/Phe/ m-Tyr Trp/Phe Trp m-Tyr m-Tyr m-Tyr m-Tyr Trp Trp m-Tyr Tyr
HC=CH HC=CH HC=CH HC=CH HC=CH H 2 C-CH2 HC=CH H 2 C-CH2 HC=CH HC=CH HC=CH HC=CH
L2
GlyU
OH
TIC 3
O
O
R
R
Liposidomycin C: Caprazamycin A: Muraminomicin F: A-90289A:
R 1 =OH R 1 =OH R 1 =H R 1 =OSO3 H
n=6 n=6 n=5 n=6
R3 =H R3 =L1 R3 =L2 R3 =L1
R2 =OSO3 H R2 =OH R2 =H R2 =OH
O OH NH
OH 3''
NH O
R NH O
N
O
O H
H
CarU
O O
G2
R
A-503083A: A-102395: A-500359A :
TIC, tetradehydro-3-isoquinoline carboxylic acid; Met , methionine sulfoxide.
G1
O
2''
R1 =CONH2 R1 =H R1 =H
R 2 =G1 R 2 =G2 R 2 =G1
O
OH O
NH
OH
n
HO R
HN O AcHN
HO HO
T1
O
NH2 O
NH2
O
O
N
O
n n
O HO
HN
T2
N
N H
N
Tunicamycins T1 T3 T1 T1 T3
n=7 n=9 n=9 n=10 n=12
HN
N
O
NH2 NH2
O
N H HO
OH
CH2OH
N
NH2
N O
Gougeron
O
Blascidin S
T3
H N
OH
OH
OH
I(Tun 13:1): III(Tun 14:1B ): V(Tun 15:1 B ): VII(Tun 16:1 A ): IX(Tun 17:1 A ):
NH2
O
OH
O
O
C
II(Tun 14:1A ): IV(Tun 15:1 A ): VI(Tun 15:0): VIII(Tun 16:1 B ): X(Tun 17:1 B ):
Amosamine
T1 T3 T2 T3 T1
n=8 n=10 n=9 n=11 n=11
N HO HO
Cytosine
O Amicetose O O O
N N
H N H N O
NH2
N H
HN
CH2OH
O
O α-Methyl-serine PABA
HO
OH
N
O NH
HO
NH2
O N O
NH2
Amicen
Mildiomycin TRENDS in Microbiology
Figure 1. Chemical structures of representative nucleoside antibiotics. Parts shaded in pink indicate the nucleoside moieties. CarU, 50 -carbamoyluridine; DABA, 2,3diaminobutyric acid; GlyU, 50 -C-glycyluridine; PABA, p-aminobenzoic acid.
and 2,3-diaminobutyric acid (DABA), other amino acids are thought to originate from primary metabolism. DABA is postulated to be synthesized from L-threonine and ammonia [6]. m-Tyr is generated from L-phenylalanine by a novel iron (II)-dependent phenylalanine-3-hydroxylase (PacX) [9]. These amino acids are linked by nine proteins that constitute the non-ribosomal peptide synthetase (NRPS) assembly line [10]. One exceptional feature of this assembly is that it does not start at the N-terminal residue AA1 and proceed to the C-terminal residue AA5. Instead, assembly is initiated by the activation and methylation of the core residue DABA and the chain is then built from the middle outward in both directions [11]. It should be noted that the transferase PacB is responsible for the transfer of the alanyl residue from alanyl-tRNA to the N terminus of the tetrapeptide intermediate yielding the pentapeptide scaffold [12]. This enzyme is unusual compared with typical NRPSs in that it catalyzes peptide bond formation in a tRNA-dependent way, hijacking an aminoacyl-tRNA from the primary metabolic pathway to the secondary antibiotic biosynthetic pathway. Rebecca Goss and colleagues have shown that the biogenesis of the unique 30 -deoxy-40 ,50 -enamino-uridine proceeds through three steps (Figure 2A). Uridine is first oxidized by flavindependent dehydrogenase (Pac11/PacK) to uridine-50 -aldehyde (UA), which is then subjected to 30 ,40 -dehydration
and 50 -transamination through the action of a Cupin family enzyme (Pac13/PacM) and pyridoxal-50 -phosphate (PLP)-dependent aminotransferase (Pac5/PacE) [13]. The free-standing condensation protein PacI is thought to then catalyze the release of the assembled peptide and its linkage with the nucleoside scaffold to build pacidamycins [10]. Six genes (lipK, lipL, lipO, lipP, lipM, and lipN) are essential for the formation of the aminoribosyl moiety in lipopeptide antibiotics (Figure 2B). The pathway is initiated by oxidative dephosphorylation of uridine-50 monophosphate (UMP) to UA, catalyzed by a non-heme iron (II)-dependent dioxygenase (LipL) [14]. This is in contrast to the first step in the biosynthesis of the pacidamycin nucleoside (30 -deoxy-40 ,50 -enamino-uridine), where uridine is oxidized to UA. Subsequently, UA is converted to 50 -amino-50 -deoxyuridine through the action of the L-methionine:UA aminotransferase (LipO). 50 -Amino-50 deoxyuridine then serves as the substrate for a phosphorylase (LipP) to generate 50 -amino-50 -deoxy-a-D-ribose-1phosphate. This sugar-1-phosphate is then processed by two enzymes in a manner that parallels typical glycosylation events: the 5-amino-5-deoxy-a-D-ribose-1-phosphate is activated by a nucleotidylyltransferase (LipM) as the nucleotide-50 -diphosphate (NDP)-sugar [uridine-50 -diphosphate (UDP)-50 -amino-50 -deoxy-a-D-ribose] and the sugar 111
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
H N
O
(A)
H N
O
O
O
O N
O
HO
HO
Pac11 H HO
OH
(B)
HO
HO
N
O
O
H N
OH
ONDP
HO
OH
O
O N
O
H
HO
OH
HO
LipM
OPO
OH
O
LipL
N
O
O
Lip PH N
NTP PPi 5'-Amino-5'-deoxyuridine 5'-Amino-5'-deoxy-αD-ribose-1-phosphate LipO H
H N
O
OH
3'-Deoxy-4', 5'enamino-uridine
O
N
O
H N
Pacidamycin Napsamycin Mureidomycin Sansanmycin
N
O
OH
H N
O
H N
Pac5
N
O
H
UA O
H O OP
Pac13
OH
Uridine
O
O
N
O
H N
O
H N
O
UMP
NDP
OH
UA
Gly
LipK
OH OH HN O O
HO
HO
H N
H N
H N
O
O
LipN
N
HO
OH
O
O
HN
O O
Liposidomycin Caprazamycin A-90289A
N
HO
OH
O
OH
5'-C-Glycyluridine (C) OH NH
NH
O O
HO
OH
O
OH
O
O
O
N
O O CapS
H
O O
NH SAM
O
O
OH OH NH2 O O O H O
SAH
NH O N
O
CapW
O O
O
NH 2
O O
O
O
HN
H N
OH NH 2 O O O H
HN
NH N
O
O
NH 2
Methanol O
NH2
A-503083 B L-Aminocaprolactam
NH 2
(D)
N N H 2 O 3 OP
O
CMP hydrolase (BlsM)
O
NH 2
N H
OH OH CMP
UDP-glucuronic acid
N O
CGA synthase (BlsD,GouF)
O HO HO
OH O OH CGA
Cytosine
NH 2 N
N O
Blascidin S Arginomycin Gougeron Ningnanmycin
CMP hydroxymethylase (MilA) NH 2 HOH 2 C
N N
H 2O 3O P
O OH
NH 2 O
HOH 2 C
UDP-glucuronic acid
N
N O CMP hydrolase CGA synthase H (MilB) (MilC) Hydroxymethyl cytosine
O HO HO
OH CH 2 OH NH 2 O N N OH O HM-CGA
Mildiomycin
HM-CMP TRENDS in Microbiology
Figure 2. Representative pathways for biosynthesis of nucleoside antibiotics. (A) Biosynthesis of nucleoside skeleton from uridine in uridyl peptide antibiotics. (B) Biosynthesis of aminoribosyl moiety from uridine-50 -monophosphate (UMP) in lipopeptide antibiotics. (C) The tandem reactions catalyzed by CapS and CapW in the amide bond formation of A-503083 B. (D) Biosynthetic pathway for the formation of cytosylglucuronic acid (CGA) and hydroxymethyl-CGA (HM-CGA) from cytidine-50 monophosphate (CMP) in cytosine-derived nucleoside antibitics. HM-CMP, hydroxymethyl-CMP; NDP, nucleoside diphosphate; NTP, nucleoside triphosphate; SAH, Sadenosylhomocysteine; SAM, S-adenosylmethionine; UA, uridine-50 -aldehyde. Adapted from [13,15,20,35].
112
Review is transferred by a glycosyltransferase (LipN) to an acceptor molecule to generate the final disaccharide [15]. The potential acceptor for UDP-5-amino-5-deoxy-a-D-ribose is 50 -C-glycyluridine, which is generated by a PLP-dependent L-threonine:uridine-50 -aldehyde transaldolase, LipK [16]. The rest of the biosynthetic steps, including the formation of the diazepanone ring and attachment of the fatty acid, remain elusive. The minimal gene cluster for tunicamycin biosynthesis has been identified [17,18] and TunA and TunF were characterized respectively as a dehydratase and an epimerase, participating in the formation of a unique enol ether [UDP-6-deoxy-5,6-ene-N-acetylgalactosaintermediate mine (GalNAc)] from UDP-GlcNAc and uridine [19]. A deeper understanding of the rest of the enzymatic processes may offer the opportunity to alter this unique natural product scaffold and reduce its cytotoxicity to mammalian cells. The biosynthetic gene clusters for A-500359s and A-503083s have been reported [20,21]. CapP is an ATPdependent capuramycin phosphotransferase that transfers the g-phosphate of ATP to the 30 -hydroxyl of the unsaturated hexuronic acid moiety of A-503083 B and confers selective resistance to A-500359 B [22]. CapS functions as an S-adenosylmethionine (SAM)-dependent carboxyl methyltransferase and activates the carboxylic acid substrate to the methyl ester at the expense of SAM. In the presence of free L-aminocaprolactam, CapW converts the methyl ester to A-503083 B by generating a new amide bond (Figure 2C). The tandem reactions catalyzed by CapS and CapW represent an ATP-independent strategy for carboxylic acid activation and intermolecular amide bond formation [20], a departure from the widespread mechanism of amide bond formation that requires the hydrolysis of ATP [23]. Biosynthesis of antifungal nucleoside antibiotics: nikkomycin, polyoxin, blasticidin S, arginomycin, and mildiomycin Nikkomycin and polyoxin are potent competitive inhibitors of fungal chitin synthases since their chemical structures are similar to the natural substrate of chitin synthase, UDPN-acetylglucosamine (Figure 3A,B). They are nontoxic to mammals, making them valuable antifungal agents. Nikkomycins comprise a di- or tripeptide backbone containing the non-proteinogenic amino acid residue hydroxypyridylhomothreonine (HPHT) and an aminohexuronic acid that is N-glycosidically linked to a uracil (nikkomycin Z and J) or 4-formyl-4- imidazolin-2-one (nikkomycin X and I) base (Figure 3A). Developed as an orphan drug for the treatment of fungal infections, nikkomycin Z has been used in Phase I clinical trials for the treatment of coccidioidomycosis [24]. It is also effective in treating naturally acquired coccidioidomycosis in dogs and ecologically significant Batrachochytrium dendrobatidis infection in frogs [25,26]. Polyoxin differs from nikkomycin mainly in the peptidyl moiety with two unusual amino acid residues, polyoximic acid (POIA) and carbamoylpolyoxamic acid (CPOAA) (Figure 3B). The biosynthetic pathways of nikkomycin and polyoxin were summarized by Winn et al. in 2010 [2]. Two recent studies suggested that an enolpyruvyl transferase (NikO) and aminotransferase (NikK) participate in the conversion of UMP
Trends in Microbiology February 2015, Vol. 23, No. 2
to the aminohexuronic acid [27,28], which is postulated to be linked via peptide bonds with the peptidyl moieties of nikkomycin and polyoxin. A putative amide synthetase (SanS) was proposed to be responsible for the peptide bond formation in nikkomycin biosynthesis by Streptomyces ansochromogenes [29] and its homolog (PolG) was suggested to assemble the three moieties in polyoxin biosynthesis by Streptomyces cacaoi subsp. asoensis [30]. However, the enzymatic mechanism of peptide formation remains unclear. Blasticidin S, arginomycin, and mildiomycin feature a glucuronic acid-derived hexose coupled to cytosine or hydroxymethyl cytosine as the nucleoside moiety. The peptidyl moiety in blasticidin S is the N-methyl b-arginine, while a b-methylarginine is present in arginomycin. The peptidyl moiety in mildiomycin contains a 5-guanidino-2,4dihydroxyvalerate side chain and a serine residue (Figure 1). The earliest step in blasticidin S biosynthesis is the specific coupling of UDP-glucuronic acid with free cytosine to form cytosylglucuronic acid (CGA) by CGA synthase (BlsD) (Figure 2D). UDP-glucuronic acid was suggested to originate from glucose and cytosine is generated from the hydrolysis of cytidine-50 -monophosphate (CMP) by CMP hydrolase (BlsM) [31]. The biosynthetic pathway of arginomycin is generally similar to that of blasticidin S except that two specific enzymes, an aspartate aminotransferase (ArgM) and a SAM-dependent methyltransferase (ArgN), are required for generation of b-methylarginine from L-arginine in arginomycin biosynthesis [32]. In mildiomycin biosynthesis, the pathway for hydroxymethyl-CGA (HM-CGA) biosynthesis is essentially the same as that for CGA formation in blasticidin S biosynthesis (Figure 2D). The only difference is that an additional thymidylate synthase-like enzyme (MilA) is required to catalyze the hydroxymethylation of CMP to form hydroxymethyl-CMP (HM-CMP) [33]. Homologs of BlsM (MilB) and BlsD (MilC) are respectively required for hydrolysis of HM-CMP and its subsequent condensation with UDP-glucuronic acid [34,35]. The 5-guanidino-2,4-dihydroxyvalerate side chain is derived from L-arginine and coupled to the pyranoside through an unusual C–C bond [35]. However, the mechanism for the formation of this unique structure remains unknown. Biosynthesis of antiviral nucleoside antibiotics: gougerotin, ningnanmycin, and amicetin Gougerotin comprises a CGA-derived nucleoside skeleton and a sarcosyl-D-serine dipeptide skeleton (Figure 1). Amicetin contains a disaccharide (amosamine and amicetose) linked to cytosine, p-aminobenzoic acid (PABA), and the terminal (+)-a-methylserine moieties (Figure 1). Ningnanmycin, a stereoisomer of gougerotin, contains L-serine in the dipeptide moiety (Figure 1) and has been developed as a plant virucide commercially available in China [36]. Biosynthetic studies of gougerotin were initiated recently by cloning and functional characterization of its gene cluster [37]. The formation of CGA is the same as in blasticidin S biosynthesis (Figure 2D). Surprisingly, no homolog of the BlsM CMP hydrolase was identified within the gougerotin gene cluster and the origin of the cytosine in gougerotin remains a mystery. CGA is then oxidized and aminated to obtain 4-amino-CGA, which can accept activated serine 113
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
(A)
CH
O
(C)
COR
R3
O
COR 2
O
R1
R HO
O
H 2N
N H OH
N
O
N H
NH
OH OH
OH
HPHT
NH 2
OH
CPOAA
R2
R1
O
R1
R2
R3
OH
OH
OH
H
OH
H
Glu
OH
Glu
OH
OH
CHO
CHO
NH
NH
Polyoxin N:
OH
Nikkomycin X: O
N
O
N
CHO
O
NH
Polynik A:
NH
Nikkomycin Z: N
O
N
OH O
O CHO NH
H C
NH
Nikkomycin I:
Glu
O
N
Polyoxin P: O
N
O
CHO NH
NH
Glu
Nikkomycin J: N
HOOC
OH
O
(B)
Polyoxin D:
H N
Nikkoxin B:
O
O
O
O
N O
N H OH
NH
COOH
O
O
N
O
NH
Nikkoxin C:
NH OH
OH
N
CPOAA CHO
O
COOH
O
NH
COOH NH
POIA OH
O
Polyoxin H:
H N
C
O
O
Nikkoxin D:
N N
O
N
OH
O
N
POIA
O
N H OH
O
NH OH
OH
CPOAA TRENDS in Microbiology
Figure 3. Chemical structures of nikkomycins, polyoxins, and related compounds. (A) Chemical structures of nikkomycins. (B) Chemical structures of representative polyoxins. (C) Chemical structures of novel polyoxin analogs and nikkomycin–polyoxin hybrids. CPOAA, carbamoylpolyoxamic acid; HPHT, hydroxypyridylhomothreonine; POIA, polyoximic acid.
and sarcosine sequentially to form yunnanmycin. An acylcoenzyme A (CoA) N-acyltransferase (GouJ) and an acylCoA synthetase (GouK) are suggested to be responsible for the activation and transfer of serine and sarcosine to 4-amino-CGA. However, further studies are needed to clarify their functions in the formation of two amide bonds. The final generation of gougerotin is achieved after the amidation of yunnanmycin at C-6 of glucuronic acid [37]. As in the biosynthesis of blasticidin S and mildiomycin, an early step of amicetin biosynthesis may require hydrolysis of CMP to provide free cytosine, which is coupled with amicetose to form amicetosyl-cytosine. Amosamine is then transferred to amicetosyl-cytosine to form cytosamine by the action of AmiG, forming a characteristic a-(1!4)-glycoside bond between amosamine and amicetose [38]. Next, activated PABA-CoA is attached to cytosamine to form plicacetin. Final-product amicetin is generated after the incorporation of (+)-a-methylserine [39]. AmiF was proposed to catalyze the formation of the amide bond between cytosine and PABA, and AmiR was proposed to be responsible for the formation of the amide bond between PABA 114
and (+)-a-methylserine [39]. However, biochemical evidence is required for characterization of their roles in the formation of the two amide bonds. Biosynthesis of complex nucleoside antibiotics proceeds via multiple steps from simple building blocks of nucleosides, amino acids, sugars, and fatty acids. The nucleoside is modified to constitute the core skeleton that distinguishes the different groups of nucleoside antibiotics. By the action of specialized enzymes, the nucleoside skeleton is then linked with the other components to form complex nucleoside antibiotics. Control of the biosynthesis of nucleoside antibiotics The onset and level of production of each antibiotic is subject to complex control by regulators at different hierarchical levels. At the lower level, cluster-situated regulators (CSRs) are located within the antibiotic biosynthetic clusters and modulate the transcription of neighboring biosynthetic genes. Higher-level pleiotropic regulators are situated outside biosynthetic gene clusters and affect the production of multiple antibiotics, sometimes also
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
affecting morphological development [40]. Global regulators are scattered throughout the chromosome and control both central metabolic genes and pleiotropic regulatory genes or CSR genes [41]. The regulatory mechanisms for nikkomycin, polyoxin, gougerotin, sansanmycin, and muraymycin production have been studied, while nothing is known about the regulation of biosynthesis of the other nucleoside antibiotics.
at a binding site that overlaps with AdpA-L-binding site I [45]. Control of sanG transcription therefore probably involves interplay between SabR and AdpA-L. Polyoxin biosynthesis in S. cacaoi subsp. asoensis is controlled by two CSRs (PolR and PolY), both of which are large SARPs. PolR activates polyoxin production by directly binding to the promoter regions of polC and polB [46]. The transcription of polR itself is activated by PolY. The affinity of PolY for a typical SARP-binding sequence in the promoter region of polR is enhanced by binding of ADP/ ATPgS to the ATPase domain of PolY, triggering the oligomerization of PolY [47]. Thus, by sensing ADP/ATP concentrations the ATPase domain of PolY may modulate its DNA-binding activity.
Cascade regulation of nikkomycin and polyoxin biosynthesis Streptomyces antibiotic regulatory proteins (SARPs) are the most frequently encountered CSRs. They fall into three categories: small [up to 400 amino acids (aa)], medium (401–799 aa), and large (800 aa or more) [41]. Large SARPs have three major functional domains: an N-terminal SARP domain, a central AAA domain (ATPase associated with diverse cellular activities), and a conserved C-terminal domain of unknown function [41]. Within the nikkomycin biosynthetic gene cluster in S. ansochromogenes, sanG encodes the only CSR. SanG is a large SARP that positively regulates nikkomycin biosynthesis by directly activating the diverging transcription of two adjacent transcriptional units (sanO-V and sanN-I), while the transcription of sanF-X is independent of SanG [42,43]. Initiation of sanG transcription is in turn controlled by a pleiotropic regulator, AdpA-L [44]. The role of AdpA-L in sanG transcription is complicated by the presence of five AdpA-L-binding sites (I–V) in the upstream region of sanG; two (I and V) are used for activation of sanG transcription, while the other three (II, III, and IV) lead to repression (Figure 4). In addition, AdpA-L exerts its effect on the transcription of sanF-X through an unknown mediator [44]. Involvement of a hormone-like autoregulator in nikkomycin biosynthesis was also indicated: SabR, a homolog of A-factor receptor protein (ArpA) in S. ansochromogenes, can also activate sanG transcription by directly binding its upstream region
Fine-tuning regulation of gougerotin biosynthesis Within the gougerotin biosynthetic gene cluster, gouR encodes the only CSR. GouR belongs to the TetR family of regulators [37]. Unlike SARP family regulators, which serve as activators, most TetR family regulators are repressors, although a few are activators [48]. GouR plays a dual role in regulation of gougerotin biosynthesis [49]. It can inhibit the transcription of gouL–gouB and thereby reduce intracellular gougerotin. It can also activate the transcription of gouM and export more gougerotin outside the cell. The delicate regulation of GouR ensures fine-tuning of the gougerotin biosynthesis and export process [49]. In most cases, TetR-like regulators involved in antibiotic production can sense small-molecule ligands (antibiotic end products and/or their biosynthetic intermediates), which modulate their DNA-binding activity. For example, during biosynthesis of the aromatic polyketide actinorhodin in Streptomyces coelicolor, the TetR-family regulator ActR binds both actinorhodin and its biosynthetic intermediate, relieving ActR-mediated repression of efflux genes (actAB) and triggering initial expression of export pumps before the final product is synthesized in large amounts. Sustained
GBL SabR
SabR AdpA-L
? SanG
I
II III IV
V
sanG mRNA
V U
T
S R
Q P
O
N M
L
K
J
H
I
F
A
B
C
D
mRNAs
mRNAs
mRNAs
Enzymes
Enzymes
Enzymes
X
SanG
Nikkomycin producon TRENDS in Microbiology
Figure 4. Cascade regulation of nikkomycin biosynthesis. SanG controls nikkomycin biosynthesis through directly binding to the bidirectional sanN–sanO promoter region and activating the transcription of the biosynthetic genes. AdpA-L switches on nikkomycin biosynthesis via activating the transcriptional initiation of sanG and sanF-X. Five AdpA-L-binding sites are located in the upstream region of sanG. AdpA-L-binding sites I and V positively control the transcription of sanG, whereas sites II, III, and IV appear to repress the activity of the sanG promoter. SabR modulates nikkomycin biosynthesis as an enhancer via interaction with the promoter region of sanG and g-butyrolactone (GBL) as a signal molecule.
115
Review expression of the actinorhodin efflux pumps requires the participation of the actinorhodin final product [50]. In another example (jadomycin biosynthesis in Streptomyces venezuelae), the TetR-family regulator JadR* can respond to early intermediates (2,3-dehydro-UWM6 and dehydrorabelomycin) and release the repression of jadY to ensure the timely supply of cofactors for a later stage in jadomycin biosynthesis [51]. However, nothing is known about the signal that triggers GouR to exert its regulatory functions. Feedback regulation of uridyl peptide antibiotics In sansanmycin production, the CSR SsaA contains an N-terminal fork head-associated (FHA) domain and a C-terminal LuxR-type helix–turn–helix DNA-binding domain (DBD). It functions as an activator of sansanmycin biosynthesis by binding to five different regions within the sansanmycin gene cluster. The DNA-binding activity of SsaA is inhibited by sansanmycins A and H in a concentration-dependent manner [8]. It seems that SsaA controls the production of sansanmycins in a feedback regulatory mechanism. A similar end product-mediated feedback mechanism was observed with JadR1, an atypical response regulator in jadomycin biosynthesis by S. venezuelae [52]. Typical FHA domains act as phosphorylation-dependent protein–protein interaction modules that preferentially bind to phosphothreonine residues in their targets. However, the conserved amino acid residues important for the binding of phosphothreonine are substituted in SsaA [8], suggesting that SsaA may respond to signals in a phosphorylation-independent manner. Highly conserved SsaA homologs (>80% identity) were found in Streptomyces coeruleorubidus (PacA), Streptomyces sp. DSM5940 (NpsM), and Streptomyces roseosporus NRRL15998 (SrosN15). Although analysis of the genome sequence suggested that S. roseosporus NRRL 15998 has the potential to produce napsamycin, it has not been shown to do so [7]. There is no report on the roles of the SsaA homologs in the control of pacidomycin and napsamycin production, but the highly conserved SsaA-binding sites were also detected in four different regions within the pacidomycin and napsamycin gene clusters [8], suggesting that end product-mediated feedback regulation is general in the production of uridyl peptide antibiotics. An SsaA homolog with lower identity (47%) was also found in Streptomyces sp. NRRL 30471 (Mur33). Similar to SsaA, Mur33 functions as an activator in muraymycin biosynthesis [53], but the regulatory mechanism was not investigated in detail. It will be valuable to investigate both the molecular basis of SsaA–ligand interactions and their broader significance in relation to antibiotic production. Further examination of SrosN15 in S. roseosporus NRRL15998 shows that two conserved amino acid residues of SsaA and NpsM were replaced by Ala55 and Gly76 in the FHA domain of SrosN15, while the DBD domain remains essentially the same. It would be interesting to examine the effect on napsamycin production of amino acid substitutions at Ala55 and Gly76 in SrosN15. Different regulatory strategies have been discovered in pathways for the biosynthesis of nucleoside antibiotics. The general finding is that different signaling molecules are implicated in the regulation of biosynthesis of nucleoside 116
Trends in Microbiology February 2015, Vol. 23, No. 2
antibiotics. Signal inputs from these molecules may be transmitted by different levels of regulator, but they all converge on CSRs to exert their regulatory functions. Application of genetic engineering approaches in nucleoside antibiotics Based on knowledge from extensive biosynthetic studies, considerable efforts have been directed toward structurebased rational design of new antibiotics by genetic manipulation of the producers. Systematic structure–activity relationship (SAR) studies have the potential to expand the antimicrobial spectrum of nucleoside antibiotics and assist the design of potent antimicrobial inhibitors. Genetic manipulation of gene clusters The most commonly used method focuses on genetic manipulations of key structural and regulatory genes [54]. As antibiotic biosynthetic genes are clustered together, amplifications of the entire biosynthetic gene cluster may improve the production of antibiotics due to a genedosage effect. Thus, duplication of the entire nikkomycin gene cluster significantly increased nikkomycin production by S. ansochromogenes [55] and gougerotin titer was improved by combining duplication of the gene cluster with promoter engineering of key structural genes [56]. Tandem repeats of antibiotic gene clusters can be achieved by application of a site-specific recombination system based on a site-specific relaxase (ZouA) and the oriT-like recombination sites RsA and RsB [57]. Combinatorial biosynthesis Combinatorial biosynthesis takes advantage of structural similarities between two or more nucleoside antibiotics to generate hybrids. When the polyoxin biosynthetic gene cluster was introduced into a nikkomycin nonproducing mutant of S. ansochromogenes, two hybrid antibiotics (polynik A and polyoxin N) and two novel polyoxin analogs (polyoxin O and polyoxin P) were generated [29,58]. Both polynik A and polyoxin N contain the nucleoside moiety from nikkomycin X and the peptidyl moiety (CPOAA) from polyoxin (Figure 3C). When the nikkomycin biosynthetic gene cluster was introduced into a polyoxin nonproducing mutant of Streptomyces aureochromogenes, polyoxin N and three other hybrid antibiotics (nikkoxin B–D) were produced [59]. The tripeptidyl nikkoxin B and C are analogs of nikkomycin I and J with the replacement of HPHT by CPOAA and nikkomxin D is an analog of nikkoxin B with POIA instead of glutamic acid bound to C-60 of the aminohexuronic acid (Figure 3C). Pathway engineering Precursor-directed biosynthesis and mutasynthesis have become useful tools in the generation of new antibiotic derivatives. Variations in the different components of uridyl peptide antibiotics suggest that the NRPSs responsible for the assembly of the peptide backbone have a high degree of flexibility for selection of amino acid substrates [60]. The substrate promiscuity of NRPSs responsible for the diversity of these nucleoside antibiotics [61] can also be explored for the production of new derivatives by exogenous addition of alternative amino acids [60]. Mutasynthesis requires the
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
Microorganisms producing nucleoside anbiocs
Marine microorganisms Uncultured terrestrial microorganisms
Terrestrial microorganisms
Cloning of gene cluster
Crypc gene cluster
Biosynthec pathway
Acvaon of crypc gene cluster
Pathway engineering Mutasynthesis Combinatorial biosynthesis
Novel analogs
Biochemical characterizaon
Novel nucleoside anbiocs TRENDS in Microbiology
Figure 5. Strategies for the discovery of novel nucleoside antibiotics.
generation of a mutant deficient in the formation of key biosynthetic intermediates, so that alternative intermediates fed to the mutant can be incorporated efficiently and the original end product eliminated. Two novel nikkomycin analogs (nikkomycin Px and Pz) were generated by feeding a sanL mutant of S. ansochromogenes with nicotinic acid [62]. In addition, heterologous expression of gene sets in an alternative host is a useful way to enhance the diversity of nucleoside antibiotics [30,63,64], most likely due to the presence or absence of a specific modifying enzyme in the heterologous host. Concluding remarks Novel antibiotics are urgently needed to tackle the growing threat of antibiotic-resistant pathogens. The cell wall biosynthetic protein MraY is a promising target for the discovery of new antibacterial agents. A large family of MraYinhibiting uridine-based nucleoside antibiotics exerts potent inhibitory activity against the highly refractory pathogen P. aeruginosa and/or multidrug-resistant M. tuberculosis. A thorough understanding of their biosynthetic pathways and studies on systematic SARs are required to generate analogs of MraY inhibitors for biological evaluation of their utility as anti-Pseudomonas and antimycobacterial agents. Microorganisms from unusual habitats, such as marine microorganisms and uncultured terrestrial microorganisms, represent untapped sources for the discovery of new nucleoside antibiotics (Figure 5). However, most of these microorganisms remain unculturable under standard laboratory conditions. One efficient way to tap this potential is PCR screening of environmental DNA or of libraries of strains for specific enzymes essential for the formation of a featured structure [65]. Another important source of new nucleoside antibiotics is the cryptic gene clusters embedded in microbial genomes. Various methods
are needed to wake these gene clusters and generate new chemical entities for biological evaluation [41]. Several strategies can be used to create novel nucleoside antibiotics. Mutasynthesis and combinatorial biosynthesis can be employed to generate new variants or hybrid nucleoside antibiotics (Figure 5). Advances in omics and bioinformatics have accelerated the identification of gene clusters for nucleoside antibiotics. However, functional characterization of these gene clusters is still a huge challenge and many questions remain to be answered (Box 1). Meeting this challenge requires various approaches including Box 1. Outstanding questions Apart from PacB and CapS/CapW, are there other enzymes that use an ATP-independent strategy for amide bond formation in nucleoside antibiotics? How are the nucleoside and peptidyl moieties of nikkomycin and polyoxin assembled together? What is the molecular basis of peptide bond formation between these two moieties? What is the pathway for the formation of the featured CarU in the uridyl glycosylpeptide antibiotics? What are the similarities and differences between biosynthesis of CarU in the uridyl glycosylpeptide antibiotics and biosynthesis of GlyU in the uridyl lipopeptide antibiotics? What is the molecular mechanism for the activation of two amino acid substrates and the formation of two different amide bonds in gougerotin? Furthermore, what is the molecular mechanism for the formation of two amide bonds between cytosine, PABA, and (+)-a-methylserine in amicetin biosynthesis? How does SabR interact with AdpA-L to control the transcription of SanG and subsequent nikkomycin production? What is the mediator that links the action of AdpA-L with the transcription of sanF-X? Why does the napsamycin gene cluster remain silent in Streptomyces roseosporus NRRL15998, while it is active in Streptomyces sp. DSM5940? Is it possible that the silencing of the napsamycin gene cluster is due to substitutions of conserved amino acids in SrosN15 of S. roseosporus NRRL15998? 117
Review bioinformatics, genome mining, gene disruption, heterologous expression, and biochemical analysis. Furthermore, comparative analysis of gene clusters for structure-related antibiotics will help us to elucidate biosynthetic pathways. Acknowledgments This work was supported by grants from the Ministry of Science and Technology of China (grant numbers 2013CB734001 and 2015CB150602) and the National Natural Science Foundation of China (grant numbers 31270110 and 31030003). The authors are grateful to Professor Keith Chater (John Innes Centre, Norwich, UK) for critical reading in the preparation of this review.
References 1 Isono, K. (1988) Nucleoside antibiotics: structure, biological activity, and biosynthesis. J. Antibiot. (Tokyo) 41, 1711–1739 2 Winn, M. et al. (2010) Antimicrobial nucleoside antibiotics targeting cell wall assembly: recent advances in structure–function studies and nucleoside biosynthesis. Nat. Prod. Rep. 27, 279–304 3 Kimura, K. and Bugg, T.D. (2003) Recent advances in antimicrobial nucleoside antibiotics targeting cell wall biosynthesis. Nat. Prod. Rep. 20, 252–273 4 Xie, Y. et al. (2008) Sansanmycins B and C, new components of sansanmycins. J. Antibiot. (Tokyo) 61, 237–240 5 Rackham, E.J. et al. (2010) Pacidamycin biosynthesis: identification and heterologous expression of the first uridyl peptide antibiotic gene cluster. Chembiochem 11, 1700–1709 6 Zhang, W. et al. (2010) Identification of the biosynthetic gene cluster for the pacidamycin group of peptidyl nucleoside antibiotics. Proc. Natl Acad. Sci. U.S.A. 107, 16828–16833 7 Kaysser, L. et al. (2011) Identification of a napsamycin biosynthesis gene cluster by genome mining. Chembiochem 12, 477–487 8 Li, Q. et al. (2013) SsaA, a member of a novel class of transcriptional regulators, controls sansanmycin production in Streptomyces sp. strain SS through a feedback mechanism. J. Bacteriol. 195, 2232–2243 9 Zhang, W. et al. (2011) Identification of phenylalanine 3-hydroxylase for meta-tyrosine biosynthesis. Biochemistry 50, 5401–5403 10 Zhang, W. et al. (2011) Nine enzymes are required for assembly of the pacidamycin group of peptidyl nucleoside antibiotics. J. Am. Chem. Soc. 133, 5240–5243 11 Walsh, C.T. and Zhang, W. (2011) Chemical logic and enzymatic machinery for biological assembly of peptidyl nucleoside antibiotics. ACS Chem. Biol. 6, 1000–1007 12 Zhang, W. et al. (2011) tRNA-dependent peptide bond formation by the transferase PacB in biosynthesis of the pacidamycin group of pentapeptidyl nucleoside antibiotics. Proc. Natl. Acad. Sci. U.S.A. 108, 12249–12253 13 Ragab, A.E. et al. (2011) Biogenesis of the unique 40 ,50 dehydronucleoside of the uridyl peptide antibiotic pacidamycin. J. Am. Chem. Soc. 133, 15288–15291 14 Yang, Z. et al. (2011) Characterization of LipL as a non-heme, Fe(II)dependent alpha-ketoglutarate:UMP dioxygenase that generates uridine-50 -aldehyde during A-90289 biosynthesis. J. Biol. Chem. 286, 7885–7892 15 Chi, X. et al. (2011) Biosynthetic origin and mechanism of formation of the aminoribosyl moiety of peptidyl nucleoside antibiotics. J. Am. Chem. Soc. 133, 14452–14459 16 Barnard-Britson, S. et al. (2012) Amalgamation of nucleosides and amino acids in antibiotic biosynthesis: discovery of an Lthreonine:uridine-50 -aldehyde transaldolase. J. Am. Chem. Soc. 134, 18514–18517 17 Wyszynski, F.J. et al. (2010) Dissecting tunicamycin biosynthesis by genome mining: cloning and heterologous expression of a minimal gene cluster. Chem. Sci. 1, 581–589 18 Chen, W. et al. (2010) Characterization of the tunicamycin gene cluster unveiling unique steps involved in its biosynthesis. Protein Cell 1, 1093–1105 19 Wyszynski, F.J. et al. (2012) Biosynthesis of the tunicamycin antibiotics proceeds via unique exo-glycal intermediates. Nat. Chem. 4, 539–546 20 Funabashi, M. et al. (2010) An ATP-independent strategy for amide bond formation in antibiotic biosynthesis. Nat. Chem. Biol. 6, 581–586 118
Trends in Microbiology February 2015, Vol. 23, No. 2
21 Funabashi, M. et al. (2009) Identification of the biosynthetic gene cluster of A-500359s in Streptomyces griseus SANK60196. J. Antibiot. (Tokyo) 62, 325–332 22 Yang, Z. et al. (2010) Functional and kinetic analysis of the phosphotransferase CapP conferring selective self-resistance to capuramycin antibiotics. J. Biol. Chem. 285, 12899–12905 23 Hollenhorst, M.A. et al. (2009) The ATP-dependent amide ligases DdaG and DdaF assemble the fumaramoyl-dipeptide scaffold of the dapdiamide antibiotics. Biochemistry 48, 10467–10472 24 Nix, D.E. et al. (2009) Pharmacokinetics of nikkomycin Z after single rising oral doses. Antimicrob. Agents Chemother. 53, 2517–2521 25 Holden, W.M. et al. (2014) Nikkomycin Z is an effective inhibitor of the chytrid fungus linked to global amphibian declines. Fungal Biol. 118, 48–60 26 Shubitz, L.F. et al. (2013) Efficacy of nikkomycin Z for respiratory coccidioidomycosis in naturally infected dogs. Med. Mycol. 51, 747–754 27 Oberdorfer, G. et al. (2012) Structural and functional characterization of NikO, an enolpyruvyl transferase essential in nikkomycin biosynthesis. J. Biol. Chem. 287, 31427–31436 28 Binter, A. et al. (2011) Characterization of the PLP-dependent aminotransferase NikK from Streptomyces tendae and its putative role in nikkomycin biosynthesis. FEBS J. 278, 4122–4135 29 Li, J. et al. (2011) Hybrid antibiotics with the nikkomycin nucleoside and polyoxin peptidyl moieties. Metab. Eng. 13, 336–344 30 Chen, W. et al. (2009) Characterization of the polyoxin biosynthetic gene cluster from Streptomyces cacaoi and engineered production of polyoxin H. J. Biol. Chem. 284, 10627–10638 31 Grochowski, L.L. and Zabriskie, T.M. (2006) Characterization of BlsM, a nucleotide hydrolase involved in cytosine production for the biosynthesis of blasticidin S. Chembiochem 7, 957–964 32 Feng, J. et al. (2014) Biosynthesis of b-methylarginine residue of peptidyl nucleoside arginomycin in Streptomyces arginensis NRRL 15941. Appl. Environ. Microbiol. 80, 5021–5027 33 Li, L. et al. (2008) The mildiomycin biosynthesis: initial steps for sequential generation of 5-hydroxymethylcytidine 50 -monophosphate and 5-hydroxymethylcytosine in Streptoverticillium rimofaciens ZJU5119. Chembiochem 9, 1286–1294 34 Zhao, G. et al. (2014) Structure of the N-glycosidase MilB in complex with hydroxymethyl CMP reveals its Arg23 specifically recognizes the substrate and controls its entry. Nucleic Acids Res. 42, 8115–8124 35 Wu, J. et al. (2012) Analysis of the mildiomycin biosynthesis gene cluster in Streptoverticillum remofaciens ZJU5119 and characterization of MilC, a hydroxymethyl cytosyl-glucuronic acid synthase. Chembiochem 13, 1613–1621 36 Wang, K. et al. (2010) Synthesis and antiviral activities of phenanthroindolizidine alkaloids and their derivatives. J. Agric. Food Chem. 58, 2703–2709 37 Niu, G. et al. (2013) Cloning, heterologous expression, and characterization of the gene cluster required for gougerotin biosynthesis. Chem. Biol. 20, 34–44 38 Chen, R. et al. (2013) Characterizing amosamine biosynthesis in amicetin reveals AmiG as a reversible retaining glycosyltransferase. J. Am. Chem. Soc. 135, 12152–12155 39 Zhang, G. et al. (2012) Characterization of the amicetin biosynthesis gene cluster from Streptomyces vinaceusdrappus NRRL 2363 implicates two alternative strategies for amide bond formation. Appl. Environ. Microbiol. 78, 2393–2401 40 Huang, J. et al. (2005) Cross-regulation among disparate antibiotic biosynthetic pathways of Streptomyces coelicolor. Mol. Microbiol. 58, 1276–1287 41 Liu, G. et al. (2013) Molecular regulation of antibiotic biosynthesis in Streptomyces. Microbiol. Mol. Biol. Rev. 77, 112–143 42 He, X. et al. (2010) SanG, a transcriptional activator, controls nikkomycin biosynthesis through binding to the sanN–sanO intergenic region in Streptomyces ansochromogenes. Microbiology 156, 828–837 43 Liu, G. et al. (2005) A pathway-specific transcriptional regulatory gene for nikkomycin biosynthesis in Streptomyces ansochromogenes that also influences colony development. Mol. Microbiol. 55, 1855–1866 44 Pan, Y. et al. (2009) The pleiotropic regulator AdpA-L directly controls the pathway-specific activator of nikkomycin biosynthesis in Streptomyces ansochromogenes. Mol. Microbiol. 72, 710–723
Review 45 Pan, Y. et al. (2011) SabR enhances nikkomycin production via regulating the transcriptional level of sanG, a pathway-specific regulatory gene in Streptomyces ansochromogenes. BMC Microbiol. 11, 164 46 Li, R. et al. (2009) polR, a pathway-specific transcriptional regulatory gene, positively controls polyoxin biosynthesis in Streptomyces cacaoi subsp. asoensis. Microbiology 155, 1819–1831 47 Li, R. et al. (2010) PolY, a transcriptional regulator with ATPase activity, directly activates transcription of polR in polyoxin biosynthesis in Streptomyces cacaoi. Mol. Microbiol. 75, 349–364 48 Cuthbertson, L. and Nodwell, J.R. (2013) The TetR family of regulators. Microbiol. Mol. Biol. Rev. 77, 440–475 49 Wei, J. et al. (2014) GouR, a TetR family transcriptional regulator, coordinates the biosynthesis and export of gougerotin in Streptomyces graminearus. Appl. Environ. Microbiol. 80, 714–722 50 Xu, Y. et al. (2012) A two-step mechanism for the activation of actinorhodin export and resistance in Streptomyces coelicolor. MBio 3, e00191–e212 51 Zhang, Y. et al. (2013) JadR*-mediated feed-forward regulation of cofactor supply in jadomycin biosynthesis. Mol. Microbiol. 90, 884–897 52 Wang, L. et al. (2009) Autoregulation of antibiotic biosynthesis by binding of the end product to an atypical response regulator. Proc. Natl. Acad. Sci. U.S.A. 106, 8617–8622 53 Xu, D. et al. (2013) Identification of Mur34 as the novel negative regulator responsible for the biosynthesis of muraymycin in Streptomyces sp. NRRL30471. PLoS ONE 8, e76068 54 Du, D. et al. (2013) Improvement of gougerotin and nikkomycin production by engineering their biosynthetic gene clusters. Appl. Microbiol. Biotechnol. 97, 6383–6396 55 Liao, G. et al. (2010) Cloning, reassembling and integration of the entire nikkomycin biosynthetic gene cluster into Streptomyces
Trends in Microbiology February 2015, Vol. 23, No. 2
56
57
58
59
60 61 62
63
64
65
ansochromogenes lead to an improved nikkomycin production. Microb. Cell Fact. 9, 6 Jiang, L. et al. (2013) Combined gene cluster engineering and precursor feeding to improve gougerotin production in Streptomyces graminearus. Appl. Microbiol. Biotechnol. 97, 10469–10477 Murakami, T. et al. (2011) A system for the targeted amplification of bacterial gene clusters multiplies antibiotic yield in Streptomyces coelicolor. Proc. Natl. Acad. Sci. U.S.A. 108, 16020–16025 Li, J. et al. (2012) Novel polyoxins generated by heterologously expressing polyoxin biosynthetic gene cluster in the sanN inactivated mutant of Streptomyces ansochromogenes. Microb. Cell Fact. 11, 135 Zhai, L. et al. (2012) Engineering of an industrial polyoxin producer for the rational production of hybrid peptidyl nucleoside antibiotics. Metab. Eng. 14, 388–393 Gruschow, S. et al. (2009) New pacidamycin antibiotics through precursor-directed biosynthesis. Chembiochem 10, 355–360 Xie, Y. et al. (2014) NRPS substrate promiscuity leads to more potent antitubercular sansanmycin analogues. J. Nat. Prod. 77, 1744–1748 Feng, C. et al. (2014) Novel nikkomycin analogues generated by mutasynthesis in Streptomyces ansochromogenes. Microb. Cell Fact. 13, 59 Kaysser, L. et al. (2009) Identification and manipulation of the caprazamycin gene cluster lead to new simplified liponucleoside antibiotics and give insights into the biosynthetic pathway. J. Biol. Chem. 284, 14987–14996 Li, L. et al. (2013) Streptomyces lividans blasticidin S deaminase and its application in engineering a blasticidin S-producing strain for ease of genetic manipulation. Appl. Environ. Microbiol. 79, 2349–2357 Funabashi, M. et al. (2013) Structure-based gene targeting discovery of sphaerimicin, a bacterial translocase I inhibitor. Angew. Chem. Int. Ed. Engl. 52, 11607–11611
119
Nucleoside antibiotics: biosynthesis, regulation, and biotechnology Guoqing Niu and Huarong Tan State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, No 1 Beichen West Road, Chaoyang District, Beijing 100101, China
The alarming rise in antibiotic-resistant pathogens has coincided with a decline in the supply of new antibiotics. It is therefore of great importance to find and create new antibiotics. Nucleoside antibiotics are a large family of natural products with diverse biological functions. Their biosynthesis is a complex process through multistep enzymatic reactions and is subject to hierarchical regulation. Genetic and biochemical studies of the biosynthetic machinery have provided the basis for pathway engineering and combinatorial biosynthesis to create new or hybrid nucleoside antibiotics. Dissection of regulatory mechanisms is leading to strategies to increase the titer of bioactive nucleoside antibiotics. Nucleoside antibiotics Nucleoside antibiotics are a large family of microbial natural products derived from nucleosides and nucleotides. Because nucleosides and nucleotides play essential roles in most of the fundamental cellular metabolism, nucleoside antibiotics exhibit a broad spectrum of biological activities, such as antibacterial, antifungal, antiviral, insecticidal, immunostimulative, immunosuppressive, and antitumor activities [1]. Based on their biological functions, they can be classified into three major groups. The antibacterial nucleoside antibiotics target bacterial cell wall biosynthesis: they are competitive inhibitors of bacterial phospho-N-acetylmuramyl-pentapeptide translocase (translocase I, denoted MraY), which initiates the lipid cycle of peptidoglycan biosynthesis [2]. The antifungal nucleoside antibiotics also target cell wall biosynthesis: they act as competitive inhibitors of fungal chitin synthases. They can also function as inhibitors of protein biosynthesis. The antiviral nucleoside antibiotics block protein biosynthesis by inhibiting peptidyl transferase. Over the past two decades, several excellent reviews have focused on the structure, biosynthesis, and biological activity of nucleoside antibiotics [1–3]. Here we highlight recent findings on their biosynthesis and its complex regulation and summarize progress in the use of this information to generate new nucleoside antibiotics and to increase production levels.
Corresponding author: Tan, H. ([email protected]). Keywords: nucleoside antibiotics; biosynthesis; regulation; genetic engineering. 0966-842X/ ß 2014 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.tim.2014.10.007
110
Trends in Microbiology, February 2015, Vol. 23, No. 2
Biosynthesis of nucleoside antibiotics Nucleoside antibiotics are endowed with unique structural features, suggesting the occurrence of unusual enzymatic reactions during the biosynthetic process. With the rapid development of methodology and the availability of an increasing number of gene clusters, biosynthetic studies have accelerated at both genetic and biochemical levels, providing the basis for pathway engineering and combinatorial biosynthesis of nucleoside antibiotics. Biosynthesis of antibacterial nucleoside antibiotics: pacidamycin, liposidomycin, tunicamycin, and capuramycin This group, which is also referred to as uridine-based nucleosides, can be further divided into four subfamilies: uridyl peptide antibiotics, uridyl lipopeptide antibiotics, uridyl lipodisaccharide antibiotics, and uridyl glycosylpeptide antibiotics. The uridyl peptide antibiotics encompass pacidamycin, napsamycin, mureidomycin, and sansanmycin. They share a common structural scaffold, a unique 30 -deoxy40 ,50 -enamino-uridine nucleoside linked to a pseudotetraor pentapeptide backbone (Figure 1), and exhibit selective antibacterial activity against Pseudomonas aeruginosa, a common nosocomial pathogen that is intrinsically resistant to various clinically used antibiotics. Sansanmycin also displays inhibition against multidrug-resistant Mycobacterium tuberculosis strains [4]. The uridyl lipopeptide antibiotics are represented by liposidomycin, caprazamycin, muraymycin, muraminomicin, and A-90289. They are structurally characterized by a 50 -C-glycyluridine (GlyU), an aminoribofuranoside, a diazepanone, and variable fatty acid side chains (Figure 1). The uridyl lipodisaccharide antibiotics are represented by tunicamycins comprising an unusual 11-carbon aminodialdose core, uracil, N-acetylglucosamine (GlcNAc), and variable fatty acyl moieties (Figure 1). The uridyl glycosylpeptide antibiotics include capuramycin and related compounds (A-503083, A-102395, and A-500359), which are characterized by a 50 -carbamoyluridine (CarU), an unsaturated hexuronic acid, and an aminocaprolactam ring (Figure 1). Comparison of gene clusters shows high sequence similarities of the corresponding proteins involved in the biosynthesis of pacidamycin, napsamycin/mureidomycin, and sansanmycin [5–8]. Most genetic and biochemical evidence comes from pacidamycin biosynthesis. Biosynthesis of the peptide backbone has been investigated extensively by Christopher Walsh and colleagues. Apart from the two non-proteinogenic amino acids meta-tyrosine (m-Tyr)
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
R1
AA 2
O
N
AA 5
H N
AA4
O
O N-Methyl-DABA
TIC2
m-Tyr Ala m-Tyr – m-Tyr Gly Ala/TIC 1 – TIC 2/TIC 3 – TIC 2/TIC 3 – –/Gly m-Tyr –/Gly m-Tyr m-Tyr – m-Tyr – TIC 4 – m-Tyr –
Pacidamycin 1/2/3: Pacidamycin 4/5/5T: Pacidamycin 6/7: Pacidamycin D/4N: Napsamycin A/B: Napsamycin C/D: MureidomycinA/C: MureidomycinB/D: Sansanmycin A/B/C: Sansanmycin D/E: Sansanmycin F/G: Sansanmycin H:
NH O
N
RO
O
OH
OH
AA 2
L1
n
O
NH
NH
R1
O
N
O
N
N H
O
O
X
H N
OH
H O
O
OH
TIC 1
O
H N
O
O
NH
NH NH
X
O
O
OH O
N
HN HO
TIC 4 HO
AA 4
AA 5
X-X
Ala Ala Ala Ala Met Met Met Met Met/Leu/MetSO Met/MetSO Met/Leu Met
Trp/Phe/ m-Tyr Trp/Phe/ m-Tyr Trp/Phe Trp m-Tyr m-Tyr m-Tyr m-Tyr Trp Trp m-Tyr Tyr
HC=CH HC=CH HC=CH HC=CH HC=CH H 2 C-CH2 HC=CH H 2 C-CH2 HC=CH HC=CH HC=CH HC=CH
L2
GlyU
OH
TIC 3
O
O
R
R
Liposidomycin C: Caprazamycin A: Muraminomicin F: A-90289A:
R 1 =OH R 1 =OH R 1 =H R 1 =OSO3 H
n=6 n=6 n=5 n=6
R3 =H R3 =L1 R3 =L2 R3 =L1
R2 =OSO3 H R2 =OH R2 =H R2 =OH
O OH NH
OH 3''
NH O
R NH O
N
O
O H
H
CarU
O O
G2
R
A-503083A: A-102395: A-500359A :
TIC, tetradehydro-3-isoquinoline carboxylic acid; Met , methionine sulfoxide.
G1
O
2''
R1 =CONH2 R1 =H R1 =H
R 2 =G1 R 2 =G2 R 2 =G1
O
OH O
NH
OH
n
HO R
HN O AcHN
HO HO
T1
O
NH2 O
NH2
O
O
N
O
n n
O HO
HN
T2
N
N H
N
Tunicamycins T1 T3 T1 T1 T3
n=7 n=9 n=9 n=10 n=12
HN
N
O
NH2 NH2
O
N H HO
OH
CH2OH
N
NH2
N O
Gougeron
O
Blascidin S
T3
H N
OH
OH
OH
I(Tun 13:1): III(Tun 14:1B ): V(Tun 15:1 B ): VII(Tun 16:1 A ): IX(Tun 17:1 A ):
NH2
O
OH
O
O
C
II(Tun 14:1A ): IV(Tun 15:1 A ): VI(Tun 15:0): VIII(Tun 16:1 B ): X(Tun 17:1 B ):
Amosamine
T1 T3 T2 T3 T1
n=8 n=10 n=9 n=11 n=11
N HO HO
Cytosine
O Amicetose O O O
N N
H N H N O
NH2
N H
HN
CH2OH
O
O α-Methyl-serine PABA
HO
OH
N
O NH
HO
NH2
O N O
NH2
Amicen
Mildiomycin TRENDS in Microbiology
Figure 1. Chemical structures of representative nucleoside antibiotics. Parts shaded in pink indicate the nucleoside moieties. CarU, 50 -carbamoyluridine; DABA, 2,3diaminobutyric acid; GlyU, 50 -C-glycyluridine; PABA, p-aminobenzoic acid.
and 2,3-diaminobutyric acid (DABA), other amino acids are thought to originate from primary metabolism. DABA is postulated to be synthesized from L-threonine and ammonia [6]. m-Tyr is generated from L-phenylalanine by a novel iron (II)-dependent phenylalanine-3-hydroxylase (PacX) [9]. These amino acids are linked by nine proteins that constitute the non-ribosomal peptide synthetase (NRPS) assembly line [10]. One exceptional feature of this assembly is that it does not start at the N-terminal residue AA1 and proceed to the C-terminal residue AA5. Instead, assembly is initiated by the activation and methylation of the core residue DABA and the chain is then built from the middle outward in both directions [11]. It should be noted that the transferase PacB is responsible for the transfer of the alanyl residue from alanyl-tRNA to the N terminus of the tetrapeptide intermediate yielding the pentapeptide scaffold [12]. This enzyme is unusual compared with typical NRPSs in that it catalyzes peptide bond formation in a tRNA-dependent way, hijacking an aminoacyl-tRNA from the primary metabolic pathway to the secondary antibiotic biosynthetic pathway. Rebecca Goss and colleagues have shown that the biogenesis of the unique 30 -deoxy-40 ,50 -enamino-uridine proceeds through three steps (Figure 2A). Uridine is first oxidized by flavindependent dehydrogenase (Pac11/PacK) to uridine-50 -aldehyde (UA), which is then subjected to 30 ,40 -dehydration
and 50 -transamination through the action of a Cupin family enzyme (Pac13/PacM) and pyridoxal-50 -phosphate (PLP)-dependent aminotransferase (Pac5/PacE) [13]. The free-standing condensation protein PacI is thought to then catalyze the release of the assembled peptide and its linkage with the nucleoside scaffold to build pacidamycins [10]. Six genes (lipK, lipL, lipO, lipP, lipM, and lipN) are essential for the formation of the aminoribosyl moiety in lipopeptide antibiotics (Figure 2B). The pathway is initiated by oxidative dephosphorylation of uridine-50 monophosphate (UMP) to UA, catalyzed by a non-heme iron (II)-dependent dioxygenase (LipL) [14]. This is in contrast to the first step in the biosynthesis of the pacidamycin nucleoside (30 -deoxy-40 ,50 -enamino-uridine), where uridine is oxidized to UA. Subsequently, UA is converted to 50 -amino-50 -deoxyuridine through the action of the L-methionine:UA aminotransferase (LipO). 50 -Amino-50 deoxyuridine then serves as the substrate for a phosphorylase (LipP) to generate 50 -amino-50 -deoxy-a-D-ribose-1phosphate. This sugar-1-phosphate is then processed by two enzymes in a manner that parallels typical glycosylation events: the 5-amino-5-deoxy-a-D-ribose-1-phosphate is activated by a nucleotidylyltransferase (LipM) as the nucleotide-50 -diphosphate (NDP)-sugar [uridine-50 -diphosphate (UDP)-50 -amino-50 -deoxy-a-D-ribose] and the sugar 111
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
H N
O
(A)
H N
O
O
O
O N
O
HO
HO
Pac11 H HO
OH
(B)
HO
HO
N
O
O
H N
OH
ONDP
HO
OH
O
O N
O
H
HO
OH
HO
LipM
OPO
OH
O
LipL
N
O
O
Lip PH N
NTP PPi 5'-Amino-5'-deoxyuridine 5'-Amino-5'-deoxy-αD-ribose-1-phosphate LipO H
H N
O
OH
3'-Deoxy-4', 5'enamino-uridine
O
N
O
H N
Pacidamycin Napsamycin Mureidomycin Sansanmycin
N
O
OH
H N
O
H N
Pac5
N
O
H
UA O
H O OP
Pac13
OH
Uridine
O
O
N
O
H N
O
H N
O
UMP
NDP
OH
UA
Gly
LipK
OH OH HN O O
HO
HO
H N
H N
H N
O
O
LipN
N
HO
OH
O
O
HN
O O
Liposidomycin Caprazamycin A-90289A
N
HO
OH
O
OH
5'-C-Glycyluridine (C) OH NH
NH
O O
HO
OH
O
OH
O
O
O
N
O O CapS
H
O O
NH SAM
O
O
OH OH NH2 O O O H O
SAH
NH O N
O
CapW
O O
O
NH 2
O O
O
O
HN
H N
OH NH 2 O O O H
HN
NH N
O
O
NH 2
Methanol O
NH2
A-503083 B L-Aminocaprolactam
NH 2
(D)
N N H 2 O 3 OP
O
CMP hydrolase (BlsM)
O
NH 2
N H
OH OH CMP
UDP-glucuronic acid
N O
CGA synthase (BlsD,GouF)
O HO HO
OH O OH CGA
Cytosine
NH 2 N
N O
Blascidin S Arginomycin Gougeron Ningnanmycin
CMP hydroxymethylase (MilA) NH 2 HOH 2 C
N N
H 2O 3O P
O OH
NH 2 O
HOH 2 C
UDP-glucuronic acid
N
N O CMP hydrolase CGA synthase H (MilB) (MilC) Hydroxymethyl cytosine
O HO HO
OH CH 2 OH NH 2 O N N OH O HM-CGA
Mildiomycin
HM-CMP TRENDS in Microbiology
Figure 2. Representative pathways for biosynthesis of nucleoside antibiotics. (A) Biosynthesis of nucleoside skeleton from uridine in uridyl peptide antibiotics. (B) Biosynthesis of aminoribosyl moiety from uridine-50 -monophosphate (UMP) in lipopeptide antibiotics. (C) The tandem reactions catalyzed by CapS and CapW in the amide bond formation of A-503083 B. (D) Biosynthetic pathway for the formation of cytosylglucuronic acid (CGA) and hydroxymethyl-CGA (HM-CGA) from cytidine-50 monophosphate (CMP) in cytosine-derived nucleoside antibitics. HM-CMP, hydroxymethyl-CMP; NDP, nucleoside diphosphate; NTP, nucleoside triphosphate; SAH, Sadenosylhomocysteine; SAM, S-adenosylmethionine; UA, uridine-50 -aldehyde. Adapted from [13,15,20,35].
112
Review is transferred by a glycosyltransferase (LipN) to an acceptor molecule to generate the final disaccharide [15]. The potential acceptor for UDP-5-amino-5-deoxy-a-D-ribose is 50 -C-glycyluridine, which is generated by a PLP-dependent L-threonine:uridine-50 -aldehyde transaldolase, LipK [16]. The rest of the biosynthetic steps, including the formation of the diazepanone ring and attachment of the fatty acid, remain elusive. The minimal gene cluster for tunicamycin biosynthesis has been identified [17,18] and TunA and TunF were characterized respectively as a dehydratase and an epimerase, participating in the formation of a unique enol ether [UDP-6-deoxy-5,6-ene-N-acetylgalactosaintermediate mine (GalNAc)] from UDP-GlcNAc and uridine [19]. A deeper understanding of the rest of the enzymatic processes may offer the opportunity to alter this unique natural product scaffold and reduce its cytotoxicity to mammalian cells. The biosynthetic gene clusters for A-500359s and A-503083s have been reported [20,21]. CapP is an ATPdependent capuramycin phosphotransferase that transfers the g-phosphate of ATP to the 30 -hydroxyl of the unsaturated hexuronic acid moiety of A-503083 B and confers selective resistance to A-500359 B [22]. CapS functions as an S-adenosylmethionine (SAM)-dependent carboxyl methyltransferase and activates the carboxylic acid substrate to the methyl ester at the expense of SAM. In the presence of free L-aminocaprolactam, CapW converts the methyl ester to A-503083 B by generating a new amide bond (Figure 2C). The tandem reactions catalyzed by CapS and CapW represent an ATP-independent strategy for carboxylic acid activation and intermolecular amide bond formation [20], a departure from the widespread mechanism of amide bond formation that requires the hydrolysis of ATP [23]. Biosynthesis of antifungal nucleoside antibiotics: nikkomycin, polyoxin, blasticidin S, arginomycin, and mildiomycin Nikkomycin and polyoxin are potent competitive inhibitors of fungal chitin synthases since their chemical structures are similar to the natural substrate of chitin synthase, UDPN-acetylglucosamine (Figure 3A,B). They are nontoxic to mammals, making them valuable antifungal agents. Nikkomycins comprise a di- or tripeptide backbone containing the non-proteinogenic amino acid residue hydroxypyridylhomothreonine (HPHT) and an aminohexuronic acid that is N-glycosidically linked to a uracil (nikkomycin Z and J) or 4-formyl-4- imidazolin-2-one (nikkomycin X and I) base (Figure 3A). Developed as an orphan drug for the treatment of fungal infections, nikkomycin Z has been used in Phase I clinical trials for the treatment of coccidioidomycosis [24]. It is also effective in treating naturally acquired coccidioidomycosis in dogs and ecologically significant Batrachochytrium dendrobatidis infection in frogs [25,26]. Polyoxin differs from nikkomycin mainly in the peptidyl moiety with two unusual amino acid residues, polyoximic acid (POIA) and carbamoylpolyoxamic acid (CPOAA) (Figure 3B). The biosynthetic pathways of nikkomycin and polyoxin were summarized by Winn et al. in 2010 [2]. Two recent studies suggested that an enolpyruvyl transferase (NikO) and aminotransferase (NikK) participate in the conversion of UMP
Trends in Microbiology February 2015, Vol. 23, No. 2
to the aminohexuronic acid [27,28], which is postulated to be linked via peptide bonds with the peptidyl moieties of nikkomycin and polyoxin. A putative amide synthetase (SanS) was proposed to be responsible for the peptide bond formation in nikkomycin biosynthesis by Streptomyces ansochromogenes [29] and its homolog (PolG) was suggested to assemble the three moieties in polyoxin biosynthesis by Streptomyces cacaoi subsp. asoensis [30]. However, the enzymatic mechanism of peptide formation remains unclear. Blasticidin S, arginomycin, and mildiomycin feature a glucuronic acid-derived hexose coupled to cytosine or hydroxymethyl cytosine as the nucleoside moiety. The peptidyl moiety in blasticidin S is the N-methyl b-arginine, while a b-methylarginine is present in arginomycin. The peptidyl moiety in mildiomycin contains a 5-guanidino-2,4dihydroxyvalerate side chain and a serine residue (Figure 1). The earliest step in blasticidin S biosynthesis is the specific coupling of UDP-glucuronic acid with free cytosine to form cytosylglucuronic acid (CGA) by CGA synthase (BlsD) (Figure 2D). UDP-glucuronic acid was suggested to originate from glucose and cytosine is generated from the hydrolysis of cytidine-50 -monophosphate (CMP) by CMP hydrolase (BlsM) [31]. The biosynthetic pathway of arginomycin is generally similar to that of blasticidin S except that two specific enzymes, an aspartate aminotransferase (ArgM) and a SAM-dependent methyltransferase (ArgN), are required for generation of b-methylarginine from L-arginine in arginomycin biosynthesis [32]. In mildiomycin biosynthesis, the pathway for hydroxymethyl-CGA (HM-CGA) biosynthesis is essentially the same as that for CGA formation in blasticidin S biosynthesis (Figure 2D). The only difference is that an additional thymidylate synthase-like enzyme (MilA) is required to catalyze the hydroxymethylation of CMP to form hydroxymethyl-CMP (HM-CMP) [33]. Homologs of BlsM (MilB) and BlsD (MilC) are respectively required for hydrolysis of HM-CMP and its subsequent condensation with UDP-glucuronic acid [34,35]. The 5-guanidino-2,4-dihydroxyvalerate side chain is derived from L-arginine and coupled to the pyranoside through an unusual C–C bond [35]. However, the mechanism for the formation of this unique structure remains unknown. Biosynthesis of antiviral nucleoside antibiotics: gougerotin, ningnanmycin, and amicetin Gougerotin comprises a CGA-derived nucleoside skeleton and a sarcosyl-D-serine dipeptide skeleton (Figure 1). Amicetin contains a disaccharide (amosamine and amicetose) linked to cytosine, p-aminobenzoic acid (PABA), and the terminal (+)-a-methylserine moieties (Figure 1). Ningnanmycin, a stereoisomer of gougerotin, contains L-serine in the dipeptide moiety (Figure 1) and has been developed as a plant virucide commercially available in China [36]. Biosynthetic studies of gougerotin were initiated recently by cloning and functional characterization of its gene cluster [37]. The formation of CGA is the same as in blasticidin S biosynthesis (Figure 2D). Surprisingly, no homolog of the BlsM CMP hydrolase was identified within the gougerotin gene cluster and the origin of the cytosine in gougerotin remains a mystery. CGA is then oxidized and aminated to obtain 4-amino-CGA, which can accept activated serine 113
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
(A)
CH
O
(C)
COR
R3
O
COR 2
O
R1
R HO
O
H 2N
N H OH
N
O
N H
NH
OH OH
OH
HPHT
NH 2
OH
CPOAA
R2
R1
O
R1
R2
R3
OH
OH
OH
H
OH
H
Glu
OH
Glu
OH
OH
CHO
CHO
NH
NH
Polyoxin N:
OH
Nikkomycin X: O
N
O
N
CHO
O
NH
Polynik A:
NH
Nikkomycin Z: N
O
N
OH O
O CHO NH
H C
NH
Nikkomycin I:
Glu
O
N
Polyoxin P: O
N
O
CHO NH
NH
Glu
Nikkomycin J: N
HOOC
OH
O
(B)
Polyoxin D:
H N
Nikkoxin B:
O
O
O
O
N O
N H OH
NH
COOH
O
O
N
O
NH
Nikkoxin C:
NH OH
OH
N
CPOAA CHO
O
COOH
O
NH
COOH NH
POIA OH
O
Polyoxin H:
H N
C
O
O
Nikkoxin D:
N N
O
N
OH
O
N
POIA
O
N H OH
O
NH OH
OH
CPOAA TRENDS in Microbiology
Figure 3. Chemical structures of nikkomycins, polyoxins, and related compounds. (A) Chemical structures of nikkomycins. (B) Chemical structures of representative polyoxins. (C) Chemical structures of novel polyoxin analogs and nikkomycin–polyoxin hybrids. CPOAA, carbamoylpolyoxamic acid; HPHT, hydroxypyridylhomothreonine; POIA, polyoximic acid.
and sarcosine sequentially to form yunnanmycin. An acylcoenzyme A (CoA) N-acyltransferase (GouJ) and an acylCoA synthetase (GouK) are suggested to be responsible for the activation and transfer of serine and sarcosine to 4-amino-CGA. However, further studies are needed to clarify their functions in the formation of two amide bonds. The final generation of gougerotin is achieved after the amidation of yunnanmycin at C-6 of glucuronic acid [37]. As in the biosynthesis of blasticidin S and mildiomycin, an early step of amicetin biosynthesis may require hydrolysis of CMP to provide free cytosine, which is coupled with amicetose to form amicetosyl-cytosine. Amosamine is then transferred to amicetosyl-cytosine to form cytosamine by the action of AmiG, forming a characteristic a-(1!4)-glycoside bond between amosamine and amicetose [38]. Next, activated PABA-CoA is attached to cytosamine to form plicacetin. Final-product amicetin is generated after the incorporation of (+)-a-methylserine [39]. AmiF was proposed to catalyze the formation of the amide bond between cytosine and PABA, and AmiR was proposed to be responsible for the formation of the amide bond between PABA 114
and (+)-a-methylserine [39]. However, biochemical evidence is required for characterization of their roles in the formation of the two amide bonds. Biosynthesis of complex nucleoside antibiotics proceeds via multiple steps from simple building blocks of nucleosides, amino acids, sugars, and fatty acids. The nucleoside is modified to constitute the core skeleton that distinguishes the different groups of nucleoside antibiotics. By the action of specialized enzymes, the nucleoside skeleton is then linked with the other components to form complex nucleoside antibiotics. Control of the biosynthesis of nucleoside antibiotics The onset and level of production of each antibiotic is subject to complex control by regulators at different hierarchical levels. At the lower level, cluster-situated regulators (CSRs) are located within the antibiotic biosynthetic clusters and modulate the transcription of neighboring biosynthetic genes. Higher-level pleiotropic regulators are situated outside biosynthetic gene clusters and affect the production of multiple antibiotics, sometimes also
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
affecting morphological development [40]. Global regulators are scattered throughout the chromosome and control both central metabolic genes and pleiotropic regulatory genes or CSR genes [41]. The regulatory mechanisms for nikkomycin, polyoxin, gougerotin, sansanmycin, and muraymycin production have been studied, while nothing is known about the regulation of biosynthesis of the other nucleoside antibiotics.
at a binding site that overlaps with AdpA-L-binding site I [45]. Control of sanG transcription therefore probably involves interplay between SabR and AdpA-L. Polyoxin biosynthesis in S. cacaoi subsp. asoensis is controlled by two CSRs (PolR and PolY), both of which are large SARPs. PolR activates polyoxin production by directly binding to the promoter regions of polC and polB [46]. The transcription of polR itself is activated by PolY. The affinity of PolY for a typical SARP-binding sequence in the promoter region of polR is enhanced by binding of ADP/ ATPgS to the ATPase domain of PolY, triggering the oligomerization of PolY [47]. Thus, by sensing ADP/ATP concentrations the ATPase domain of PolY may modulate its DNA-binding activity.
Cascade regulation of nikkomycin and polyoxin biosynthesis Streptomyces antibiotic regulatory proteins (SARPs) are the most frequently encountered CSRs. They fall into three categories: small [up to 400 amino acids (aa)], medium (401–799 aa), and large (800 aa or more) [41]. Large SARPs have three major functional domains: an N-terminal SARP domain, a central AAA domain (ATPase associated with diverse cellular activities), and a conserved C-terminal domain of unknown function [41]. Within the nikkomycin biosynthetic gene cluster in S. ansochromogenes, sanG encodes the only CSR. SanG is a large SARP that positively regulates nikkomycin biosynthesis by directly activating the diverging transcription of two adjacent transcriptional units (sanO-V and sanN-I), while the transcription of sanF-X is independent of SanG [42,43]. Initiation of sanG transcription is in turn controlled by a pleiotropic regulator, AdpA-L [44]. The role of AdpA-L in sanG transcription is complicated by the presence of five AdpA-L-binding sites (I–V) in the upstream region of sanG; two (I and V) are used for activation of sanG transcription, while the other three (II, III, and IV) lead to repression (Figure 4). In addition, AdpA-L exerts its effect on the transcription of sanF-X through an unknown mediator [44]. Involvement of a hormone-like autoregulator in nikkomycin biosynthesis was also indicated: SabR, a homolog of A-factor receptor protein (ArpA) in S. ansochromogenes, can also activate sanG transcription by directly binding its upstream region
Fine-tuning regulation of gougerotin biosynthesis Within the gougerotin biosynthetic gene cluster, gouR encodes the only CSR. GouR belongs to the TetR family of regulators [37]. Unlike SARP family regulators, which serve as activators, most TetR family regulators are repressors, although a few are activators [48]. GouR plays a dual role in regulation of gougerotin biosynthesis [49]. It can inhibit the transcription of gouL–gouB and thereby reduce intracellular gougerotin. It can also activate the transcription of gouM and export more gougerotin outside the cell. The delicate regulation of GouR ensures fine-tuning of the gougerotin biosynthesis and export process [49]. In most cases, TetR-like regulators involved in antibiotic production can sense small-molecule ligands (antibiotic end products and/or their biosynthetic intermediates), which modulate their DNA-binding activity. For example, during biosynthesis of the aromatic polyketide actinorhodin in Streptomyces coelicolor, the TetR-family regulator ActR binds both actinorhodin and its biosynthetic intermediate, relieving ActR-mediated repression of efflux genes (actAB) and triggering initial expression of export pumps before the final product is synthesized in large amounts. Sustained
GBL SabR
SabR AdpA-L
? SanG
I
II III IV
V
sanG mRNA
V U
T
S R
Q P
O
N M
L
K
J
H
I
F
A
B
C
D
mRNAs
mRNAs
mRNAs
Enzymes
Enzymes
Enzymes
X
SanG
Nikkomycin producon TRENDS in Microbiology
Figure 4. Cascade regulation of nikkomycin biosynthesis. SanG controls nikkomycin biosynthesis through directly binding to the bidirectional sanN–sanO promoter region and activating the transcription of the biosynthetic genes. AdpA-L switches on nikkomycin biosynthesis via activating the transcriptional initiation of sanG and sanF-X. Five AdpA-L-binding sites are located in the upstream region of sanG. AdpA-L-binding sites I and V positively control the transcription of sanG, whereas sites II, III, and IV appear to repress the activity of the sanG promoter. SabR modulates nikkomycin biosynthesis as an enhancer via interaction with the promoter region of sanG and g-butyrolactone (GBL) as a signal molecule.
115
Review expression of the actinorhodin efflux pumps requires the participation of the actinorhodin final product [50]. In another example (jadomycin biosynthesis in Streptomyces venezuelae), the TetR-family regulator JadR* can respond to early intermediates (2,3-dehydro-UWM6 and dehydrorabelomycin) and release the repression of jadY to ensure the timely supply of cofactors for a later stage in jadomycin biosynthesis [51]. However, nothing is known about the signal that triggers GouR to exert its regulatory functions. Feedback regulation of uridyl peptide antibiotics In sansanmycin production, the CSR SsaA contains an N-terminal fork head-associated (FHA) domain and a C-terminal LuxR-type helix–turn–helix DNA-binding domain (DBD). It functions as an activator of sansanmycin biosynthesis by binding to five different regions within the sansanmycin gene cluster. The DNA-binding activity of SsaA is inhibited by sansanmycins A and H in a concentration-dependent manner [8]. It seems that SsaA controls the production of sansanmycins in a feedback regulatory mechanism. A similar end product-mediated feedback mechanism was observed with JadR1, an atypical response regulator in jadomycin biosynthesis by S. venezuelae [52]. Typical FHA domains act as phosphorylation-dependent protein–protein interaction modules that preferentially bind to phosphothreonine residues in their targets. However, the conserved amino acid residues important for the binding of phosphothreonine are substituted in SsaA [8], suggesting that SsaA may respond to signals in a phosphorylation-independent manner. Highly conserved SsaA homologs (>80% identity) were found in Streptomyces coeruleorubidus (PacA), Streptomyces sp. DSM5940 (NpsM), and Streptomyces roseosporus NRRL15998 (SrosN15). Although analysis of the genome sequence suggested that S. roseosporus NRRL 15998 has the potential to produce napsamycin, it has not been shown to do so [7]. There is no report on the roles of the SsaA homologs in the control of pacidomycin and napsamycin production, but the highly conserved SsaA-binding sites were also detected in four different regions within the pacidomycin and napsamycin gene clusters [8], suggesting that end product-mediated feedback regulation is general in the production of uridyl peptide antibiotics. An SsaA homolog with lower identity (47%) was also found in Streptomyces sp. NRRL 30471 (Mur33). Similar to SsaA, Mur33 functions as an activator in muraymycin biosynthesis [53], but the regulatory mechanism was not investigated in detail. It will be valuable to investigate both the molecular basis of SsaA–ligand interactions and their broader significance in relation to antibiotic production. Further examination of SrosN15 in S. roseosporus NRRL15998 shows that two conserved amino acid residues of SsaA and NpsM were replaced by Ala55 and Gly76 in the FHA domain of SrosN15, while the DBD domain remains essentially the same. It would be interesting to examine the effect on napsamycin production of amino acid substitutions at Ala55 and Gly76 in SrosN15. Different regulatory strategies have been discovered in pathways for the biosynthesis of nucleoside antibiotics. The general finding is that different signaling molecules are implicated in the regulation of biosynthesis of nucleoside 116
Trends in Microbiology February 2015, Vol. 23, No. 2
antibiotics. Signal inputs from these molecules may be transmitted by different levels of regulator, but they all converge on CSRs to exert their regulatory functions. Application of genetic engineering approaches in nucleoside antibiotics Based on knowledge from extensive biosynthetic studies, considerable efforts have been directed toward structurebased rational design of new antibiotics by genetic manipulation of the producers. Systematic structure–activity relationship (SAR) studies have the potential to expand the antimicrobial spectrum of nucleoside antibiotics and assist the design of potent antimicrobial inhibitors. Genetic manipulation of gene clusters The most commonly used method focuses on genetic manipulations of key structural and regulatory genes [54]. As antibiotic biosynthetic genes are clustered together, amplifications of the entire biosynthetic gene cluster may improve the production of antibiotics due to a genedosage effect. Thus, duplication of the entire nikkomycin gene cluster significantly increased nikkomycin production by S. ansochromogenes [55] and gougerotin titer was improved by combining duplication of the gene cluster with promoter engineering of key structural genes [56]. Tandem repeats of antibiotic gene clusters can be achieved by application of a site-specific recombination system based on a site-specific relaxase (ZouA) and the oriT-like recombination sites RsA and RsB [57]. Combinatorial biosynthesis Combinatorial biosynthesis takes advantage of structural similarities between two or more nucleoside antibiotics to generate hybrids. When the polyoxin biosynthetic gene cluster was introduced into a nikkomycin nonproducing mutant of S. ansochromogenes, two hybrid antibiotics (polynik A and polyoxin N) and two novel polyoxin analogs (polyoxin O and polyoxin P) were generated [29,58]. Both polynik A and polyoxin N contain the nucleoside moiety from nikkomycin X and the peptidyl moiety (CPOAA) from polyoxin (Figure 3C). When the nikkomycin biosynthetic gene cluster was introduced into a polyoxin nonproducing mutant of Streptomyces aureochromogenes, polyoxin N and three other hybrid antibiotics (nikkoxin B–D) were produced [59]. The tripeptidyl nikkoxin B and C are analogs of nikkomycin I and J with the replacement of HPHT by CPOAA and nikkomxin D is an analog of nikkoxin B with POIA instead of glutamic acid bound to C-60 of the aminohexuronic acid (Figure 3C). Pathway engineering Precursor-directed biosynthesis and mutasynthesis have become useful tools in the generation of new antibiotic derivatives. Variations in the different components of uridyl peptide antibiotics suggest that the NRPSs responsible for the assembly of the peptide backbone have a high degree of flexibility for selection of amino acid substrates [60]. The substrate promiscuity of NRPSs responsible for the diversity of these nucleoside antibiotics [61] can also be explored for the production of new derivatives by exogenous addition of alternative amino acids [60]. Mutasynthesis requires the
Review
Trends in Microbiology February 2015, Vol. 23, No. 2
Microorganisms producing nucleoside anbiocs
Marine microorganisms Uncultured terrestrial microorganisms
Terrestrial microorganisms
Cloning of gene cluster
Crypc gene cluster
Biosynthec pathway
Acvaon of crypc gene cluster
Pathway engineering Mutasynthesis Combinatorial biosynthesis
Novel analogs
Biochemical characterizaon
Novel nucleoside anbiocs TRENDS in Microbiology
Figure 5. Strategies for the discovery of novel nucleoside antibiotics.
generation of a mutant deficient in the formation of key biosynthetic intermediates, so that alternative intermediates fed to the mutant can be incorporated efficiently and the original end product eliminated. Two novel nikkomycin analogs (nikkomycin Px and Pz) were generated by feeding a sanL mutant of S. ansochromogenes with nicotinic acid [62]. In addition, heterologous expression of gene sets in an alternative host is a useful way to enhance the diversity of nucleoside antibiotics [30,63,64], most likely due to the presence or absence of a specific modifying enzyme in the heterologous host. Concluding remarks Novel antibiotics are urgently needed to tackle the growing threat of antibiotic-resistant pathogens. The cell wall biosynthetic protein MraY is a promising target for the discovery of new antibacterial agents. A large family of MraYinhibiting uridine-based nucleoside antibiotics exerts potent inhibitory activity against the highly refractory pathogen P. aeruginosa and/or multidrug-resistant M. tuberculosis. A thorough understanding of their biosynthetic pathways and studies on systematic SARs are required to generate analogs of MraY inhibitors for biological evaluation of their utility as anti-Pseudomonas and antimycobacterial agents. Microorganisms from unusual habitats, such as marine microorganisms and uncultured terrestrial microorganisms, represent untapped sources for the discovery of new nucleoside antibiotics (Figure 5). However, most of these microorganisms remain unculturable under standard laboratory conditions. One efficient way to tap this potential is PCR screening of environmental DNA or of libraries of strains for specific enzymes essential for the formation of a featured structure [65]. Another important source of new nucleoside antibiotics is the cryptic gene clusters embedded in microbial genomes. Various methods
are needed to wake these gene clusters and generate new chemical entities for biological evaluation [41]. Several strategies can be used to create novel nucleoside antibiotics. Mutasynthesis and combinatorial biosynthesis can be employed to generate new variants or hybrid nucleoside antibiotics (Figure 5). Advances in omics and bioinformatics have accelerated the identification of gene clusters for nucleoside antibiotics. However, functional characterization of these gene clusters is still a huge challenge and many questions remain to be answered (Box 1). Meeting this challenge requires various approaches including Box 1. Outstanding questions Apart from PacB and CapS/CapW, are there other enzymes that use an ATP-independent strategy for amide bond formation in nucleoside antibiotics? How are the nucleoside and peptidyl moieties of nikkomycin and polyoxin assembled together? What is the molecular basis of peptide bond formation between these two moieties? What is the pathway for the formation of the featured CarU in the uridyl glycosylpeptide antibiotics? What are the similarities and differences between biosynthesis of CarU in the uridyl glycosylpeptide antibiotics and biosynthesis of GlyU in the uridyl lipopeptide antibiotics? What is the molecular mechanism for the activation of two amino acid substrates and the formation of two different amide bonds in gougerotin? Furthermore, what is the molecular mechanism for the formation of two amide bonds between cytosine, PABA, and (+)-a-methylserine in amicetin biosynthesis? How does SabR interact with AdpA-L to control the transcription of SanG and subsequent nikkomycin production? What is the mediator that links the action of AdpA-L with the transcription of sanF-X? Why does the napsamycin gene cluster remain silent in Streptomyces roseosporus NRRL15998, while it is active in Streptomyces sp. DSM5940? Is it possible that the silencing of the napsamycin gene cluster is due to substitutions of conserved amino acids in SrosN15 of S. roseosporus NRRL15998? 117
Review bioinformatics, genome mining, gene disruption, heterologous expression, and biochemical analysis. Furthermore, comparative analysis of gene clusters for structure-related antibiotics will help us to elucidate biosynthetic pathways. Acknowledgments This work was supported by grants from the Ministry of Science and Technology of China (grant numbers 2013CB734001 and 2015CB150602) and the National Natural Science Foundation of China (grant numbers 31270110 and 31030003). The authors are grateful to Professor Keith Chater (John Innes Centre, Norwich, UK) for critical reading in the preparation of this review.
References 1 Isono, K. (1988) Nucleoside antibiotics: structure, biological activity, and biosynthesis. J. Antibiot. (Tokyo) 41, 1711–1739 2 Winn, M. et al. (2010) Antimicrobial nucleoside antibiotics targeting cell wall assembly: recent advances in structure–function studies and nucleoside biosynthesis. Nat. Prod. Rep. 27, 279–304 3 Kimura, K. and Bugg, T.D. (2003) Recent advances in antimicrobial nucleoside antibiotics targeting cell wall biosynthesis. Nat. Prod. Rep. 20, 252–273 4 Xie, Y. et al. (2008) Sansanmycins B and C, new components of sansanmycins. J. Antibiot. (Tokyo) 61, 237–240 5 Rackham, E.J. et al. (2010) Pacidamycin biosynthesis: identification and heterologous expression of the first uridyl peptide antibiotic gene cluster. Chembiochem 11, 1700–1709 6 Zhang, W. et al. (2010) Identification of the biosynthetic gene cluster for the pacidamycin group of peptidyl nucleoside antibiotics. Proc. Natl Acad. Sci. U.S.A. 107, 16828–16833 7 Kaysser, L. et al. (2011) Identification of a napsamycin biosynthesis gene cluster by genome mining. Chembiochem 12, 477–487 8 Li, Q. et al. (2013) SsaA, a member of a novel class of transcriptional regulators, controls sansanmycin production in Streptomyces sp. strain SS through a feedback mechanism. J. Bacteriol. 195, 2232–2243 9 Zhang, W. et al. (2011) Identification of phenylalanine 3-hydroxylase for meta-tyrosine biosynthesis. Biochemistry 50, 5401–5403 10 Zhang, W. et al. (2011) Nine enzymes are required for assembly of the pacidamycin group of peptidyl nucleoside antibiotics. J. Am. Chem. Soc. 133, 5240–5243 11 Walsh, C.T. and Zhang, W. (2011) Chemical logic and enzymatic machinery for biological assembly of peptidyl nucleoside antibiotics. ACS Chem. Biol. 6, 1000–1007 12 Zhang, W. et al. (2011) tRNA-dependent peptide bond formation by the transferase PacB in biosynthesis of the pacidamycin group of pentapeptidyl nucleoside antibiotics. Proc. Natl. Acad. Sci. U.S.A. 108, 12249–12253 13 Ragab, A.E. et al. (2011) Biogenesis of the unique 40 ,50 dehydronucleoside of the uridyl peptide antibiotic pacidamycin. J. Am. Chem. Soc. 133, 15288–15291 14 Yang, Z. et al. (2011) Characterization of LipL as a non-heme, Fe(II)dependent alpha-ketoglutarate:UMP dioxygenase that generates uridine-50 -aldehyde during A-90289 biosynthesis. J. Biol. Chem. 286, 7885–7892 15 Chi, X. et al. (2011) Biosynthetic origin and mechanism of formation of the aminoribosyl moiety of peptidyl nucleoside antibiotics. J. Am. Chem. Soc. 133, 14452–14459 16 Barnard-Britson, S. et al. (2012) Amalgamation of nucleosides and amino acids in antibiotic biosynthesis: discovery of an Lthreonine:uridine-50 -aldehyde transaldolase. J. Am. Chem. Soc. 134, 18514–18517 17 Wyszynski, F.J. et al. (2010) Dissecting tunicamycin biosynthesis by genome mining: cloning and heterologous expression of a minimal gene cluster. Chem. Sci. 1, 581–589 18 Chen, W. et al. (2010) Characterization of the tunicamycin gene cluster unveiling unique steps involved in its biosynthesis. Protein Cell 1, 1093–1105 19 Wyszynski, F.J. et al. (2012) Biosynthesis of the tunicamycin antibiotics proceeds via unique exo-glycal intermediates. Nat. Chem. 4, 539–546 20 Funabashi, M. et al. (2010) An ATP-independent strategy for amide bond formation in antibiotic biosynthesis. Nat. Chem. Biol. 6, 581–586 118
Trends in Microbiology February 2015, Vol. 23, No. 2
21 Funabashi, M. et al. (2009) Identification of the biosynthetic gene cluster of A-500359s in Streptomyces griseus SANK60196. J. Antibiot. (Tokyo) 62, 325–332 22 Yang, Z. et al. (2010) Functional and kinetic analysis of the phosphotransferase CapP conferring selective self-resistance to capuramycin antibiotics. J. Biol. Chem. 285, 12899–12905 23 Hollenhorst, M.A. et al. (2009) The ATP-dependent amide ligases DdaG and DdaF assemble the fumaramoyl-dipeptide scaffold of the dapdiamide antibiotics. Biochemistry 48, 10467–10472 24 Nix, D.E. et al. (2009) Pharmacokinetics of nikkomycin Z after single rising oral doses. Antimicrob. Agents Chemother. 53, 2517–2521 25 Holden, W.M. et al. (2014) Nikkomycin Z is an effective inhibitor of the chytrid fungus linked to global amphibian declines. Fungal Biol. 118, 48–60 26 Shubitz, L.F. et al. (2013) Efficacy of nikkomycin Z for respiratory coccidioidomycosis in naturally infected dogs. Med. Mycol. 51, 747–754 27 Oberdorfer, G. et al. (2012) Structural and functional characterization of NikO, an enolpyruvyl transferase essential in nikkomycin biosynthesis. J. Biol. Chem. 287, 31427–31436 28 Binter, A. et al. (2011) Characterization of the PLP-dependent aminotransferase NikK from Streptomyces tendae and its putative role in nikkomycin biosynthesis. FEBS J. 278, 4122–4135 29 Li, J. et al. (2011) Hybrid antibiotics with the nikkomycin nucleoside and polyoxin peptidyl moieties. Metab. Eng. 13, 336–344 30 Chen, W. et al. (2009) Characterization of the polyoxin biosynthetic gene cluster from Streptomyces cacaoi and engineered production of polyoxin H. J. Biol. Chem. 284, 10627–10638 31 Grochowski, L.L. and Zabriskie, T.M. (2006) Characterization of BlsM, a nucleotide hydrolase involved in cytosine production for the biosynthesis of blasticidin S. Chembiochem 7, 957–964 32 Feng, J. et al. (2014) Biosynthesis of b-methylarginine residue of peptidyl nucleoside arginomycin in Streptomyces arginensis NRRL 15941. Appl. Environ. Microbiol. 80, 5021–5027 33 Li, L. et al. (2008) The mildiomycin biosynthesis: initial steps for sequential generation of 5-hydroxymethylcytidine 50 -monophosphate and 5-hydroxymethylcytosine in Streptoverticillium rimofaciens ZJU5119. Chembiochem 9, 1286–1294 34 Zhao, G. et al. (2014) Structure of the N-glycosidase MilB in complex with hydroxymethyl CMP reveals its Arg23 specifically recognizes the substrate and controls its entry. Nucleic Acids Res. 42, 8115–8124 35 Wu, J. et al. (2012) Analysis of the mildiomycin biosynthesis gene cluster in Streptoverticillum remofaciens ZJU5119 and characterization of MilC, a hydroxymethyl cytosyl-glucuronic acid synthase. Chembiochem 13, 1613–1621 36 Wang, K. et al. (2010) Synthesis and antiviral activities of phenanthroindolizidine alkaloids and their derivatives. J. Agric. Food Chem. 58, 2703–2709 37 Niu, G. et al. (2013) Cloning, heterologous expression, and characterization of the gene cluster required for gougerotin biosynthesis. Chem. Biol. 20, 34–44 38 Chen, R. et al. (2013) Characterizing amosamine biosynthesis in amicetin reveals AmiG as a reversible retaining glycosyltransferase. J. Am. Chem. Soc. 135, 12152–12155 39 Zhang, G. et al. (2012) Characterization of the amicetin biosynthesis gene cluster from Streptomyces vinaceusdrappus NRRL 2363 implicates two alternative strategies for amide bond formation. Appl. Environ. Microbiol. 78, 2393–2401 40 Huang, J. et al. (2005) Cross-regulation among disparate antibiotic biosynthetic pathways of Streptomyces coelicolor. Mol. Microbiol. 58, 1276–1287 41 Liu, G. et al. (2013) Molecular regulation of antibiotic biosynthesis in Streptomyces. Microbiol. Mol. Biol. Rev. 77, 112–143 42 He, X. et al. (2010) SanG, a transcriptional activator, controls nikkomycin biosynthesis through binding to the sanN–sanO intergenic region in Streptomyces ansochromogenes. Microbiology 156, 828–837 43 Liu, G. et al. (2005) A pathway-specific transcriptional regulatory gene for nikkomycin biosynthesis in Streptomyces ansochromogenes that also influences colony development. Mol. Microbiol. 55, 1855–1866 44 Pan, Y. et al. (2009) The pleiotropic regulator AdpA-L directly controls the pathway-specific activator of nikkomycin biosynthesis in Streptomyces ansochromogenes. Mol. Microbiol. 72, 710–723
Review 45 Pan, Y. et al. (2011) SabR enhances nikkomycin production via regulating the transcriptional level of sanG, a pathway-specific regulatory gene in Streptomyces ansochromogenes. BMC Microbiol. 11, 164 46 Li, R. et al. (2009) polR, a pathway-specific transcriptional regulatory gene, positively controls polyoxin biosynthesis in Streptomyces cacaoi subsp. asoensis. Microbiology 155, 1819–1831 47 Li, R. et al. (2010) PolY, a transcriptional regulator with ATPase activity, directly activates transcription of polR in polyoxin biosynthesis in Streptomyces cacaoi. Mol. Microbiol. 75, 349–364 48 Cuthbertson, L. and Nodwell, J.R. (2013) The TetR family of regulators. Microbiol. Mol. Biol. Rev. 77, 440–475 49 Wei, J. et al. (2014) GouR, a TetR family transcriptional regulator, coordinates the biosynthesis and export of gougerotin in Streptomyces graminearus. Appl. Environ. Microbiol. 80, 714–722 50 Xu, Y. et al. (2012) A two-step mechanism for the activation of actinorhodin export and resistance in Streptomyces coelicolor. MBio 3, e00191–e212 51 Zhang, Y. et al. (2013) JadR*-mediated feed-forward regulation of cofactor supply in jadomycin biosynthesis. Mol. Microbiol. 90, 884–897 52 Wang, L. et al. (2009) Autoregulation of antibiotic biosynthesis by binding of the end product to an atypical response regulator. Proc. Natl. Acad. Sci. U.S.A. 106, 8617–8622 53 Xu, D. et al. (2013) Identification of Mur34 as the novel negative regulator responsible for the biosynthesis of muraymycin in Streptomyces sp. NRRL30471. PLoS ONE 8, e76068 54 Du, D. et al. (2013) Improvement of gougerotin and nikkomycin production by engineering their biosynthetic gene clusters. Appl. Microbiol. Biotechnol. 97, 6383–6396 55 Liao, G. et al. (2010) Cloning, reassembling and integration of the entire nikkomycin biosynthetic gene cluster into Streptomyces
Trends in Microbiology February 2015, Vol. 23, No. 2
56
57
58
59
60 61 62
63
64
65
ansochromogenes lead to an improved nikkomycin production. Microb. Cell Fact. 9, 6 Jiang, L. et al. (2013) Combined gene cluster engineering and precursor feeding to improve gougerotin production in Streptomyces graminearus. Appl. Microbiol. Biotechnol. 97, 10469–10477 Murakami, T. et al. (2011) A system for the targeted amplification of bacterial gene clusters multiplies antibiotic yield in Streptomyces coelicolor. Proc. Natl. Acad. Sci. U.S.A. 108, 16020–16025 Li, J. et al. (2012) Novel polyoxins generated by heterologously expressing polyoxin biosynthetic gene cluster in the sanN inactivated mutant of Streptomyces ansochromogenes. Microb. Cell Fact. 11, 135 Zhai, L. et al. (2012) Engineering of an industrial polyoxin producer for the rational production of hybrid peptidyl nucleoside antibiotics. Metab. Eng. 14, 388–393 Gruschow, S. et al. (2009) New pacidamycin antibiotics through precursor-directed biosynthesis. Chembiochem 10, 355–360 Xie, Y. et al. (2014) NRPS substrate promiscuity leads to more potent antitubercular sansanmycin analogues. J. Nat. Prod. 77, 1744–1748 Feng, C. et al. (2014) Novel nikkomycin analogues generated by mutasynthesis in Streptomyces ansochromogenes. Microb. Cell Fact. 13, 59 Kaysser, L. et al. (2009) Identification and manipulation of the caprazamycin gene cluster lead to new simplified liponucleoside antibiotics and give insights into the biosynthetic pathway. J. Biol. Chem. 284, 14987–14996 Li, L. et al. (2013) Streptomyces lividans blasticidin S deaminase and its application in engineering a blasticidin S-producing strain for ease of genetic manipulation. Appl. Environ. Microbiol. 79, 2349–2357 Funabashi, M. et al. (2013) Structure-based gene targeting discovery of sphaerimicin, a bacterial translocase I inhibitor. Angew. Chem. Int. Ed. Engl. 52, 11607–11611
119
