- Email: [email protected]
Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries
Please cite this article in press as: Timmermans and Ross, Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries, Trends in Biotechnology (2019), https://doi.org/10.1016/j.tibtech.2019.11.003
Trends in Biotechnology
Spotlight
Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries Marshall L. Timmermans1 and Avena C. Ross1,@,* The redundancy of natural product biosynthesis in microbes poses a practical challenge for discovering new antimicrobial compounds from bacteria. The recent application of clustered regularly interspaced short palindromic repeats (CRISPR) technology by Culp et al. to inactivate the production of abundant antibiotics generates a metabolic clean slate for the detection and/ or isolation of new and less plentiful antibiotics activated in mutant strains.
Bacteria are the main source of antibiotics used in the clinic today. However, conventional natural product drug discovery pipelines have been abandoned and, coupled with the increase in bacterial pathogens resistant to commonly used antibiotics, this has generated a profound challenge for modern medicine. A CRISPR-Cas9 approach [1] now shows that interrupting production of highly abundant natural antibiotic products produced by actinomycetes stimulates the biosynthesis of other natural products, potentially reinvigorating antibiotic discovery from strains discarded from conventional drug discovery pipelines. While one approach for identifying novel bioactive natural products from bacteria is to investigate new and exotic bacteria, recent insights from genome sequencing indicate that it is also worthwhile reinvestigating the untapped biosynthetic potential of previously characterized strains. Analysis
shows that actinobacterial genomes encode, on average, 13 natural product biosynthetic gene clusters (BGCs), although they might only produce one or two natural products during laboratory fermentation [2]. Stimulating the biosynthesis of new natural products arising from these so-called ‘silent’ BGCs through various physiological and genetic means has been a high priority for the research community in recent years. Typically, this is done by exposing the bacteria to simulated environmental stressors, such as extremes of temperature, pH, toxic metal ions, or foreign bacterial species [3]. Alternatively, genetic manipulation of bacterial strains is used to alter the regulation of silent BGCs. This can be done by inserting constitutive promoters ahead of uncharacterized BGCs [4], engineering of transcription factors and ribosomes to alter global transcription and translation patterns [5], or by heterologous expression of either intact or refactored BGCs [6]. Culp and colleagues recently described a genetic strategy to increase the production of new actinobacterial natural products by inactivating the biosynthetic machinery for otherwise abundant confounding characterized antibiotics (Figure 1) [1]. Several of the most commonly rediscovered antibiotics in traditional drug discovery screens, the aminoglycosides streptomycin and streptothricin [7], were chosen because they are produced by 1% and 10% of actinomycetes, respectively. Using CRISPR Cas9 methodology, targeted disruption of genes in the streptomycin or streptothricin BGCs in a series of actinomycetes capable of producing the respective antibiotics was achieved. By using highly conserved sequences within the genes of the streptomycin and streptothricin BGCs, a single set of genetic constructs can be used for high-throughput deletion of related
pathways in many strains. This approach will enable the expedited drug discovery screens of large libraries in the future without requiring individually tailored CRISPR constructs or genome sequences. Mutant and wild-type strains can then be subjected to antibiotic activity assays and chemical analysis using the Global Natural Products Social Molecular Networking pipeline (GNPS) to identify newly produced or upregulated secondary metabolites/antibiotics [8]. All knockout strains generated produce different compound profiles compared with wild-type, but, interestingly there is also variation depending on which part of the antibiotic BGC is eliminated. This supports the hypothesis that abolishing the production of abundant secondary metabolites alters the flux of metabolic precursors within the cell and/or the genetic regulatory circuits responsive to these metabolites. The success of any discovery platform is measured in the natural products identified therein. Culp and colleagues identified the new production of known compounds from their CRISPR mutants, including the thiotetronate antibiotic thiolactomycin, the halogenated amino acid derivative 5-chloro3-formyl indole, and the aglycone of the cytotoxic jadomycin derivative phenanthroviridin. Additional upregulation in the production of the rare antibiotics bamicetin and amicetin, and numerous variants of the siderophore ferrrioxamine were also observed. While the structures and antibiotic activity of these particular compounds are already well characterized, these recent results suggest that the inactivation approach is effective at both activating the expression and increasing the titer of minor natural products [1]. Additionally, abolishing the production of an abundant natural product also
Trends in Biotechnology, -- 2019, Vol. --, No. --
1
Please cite this article in press as: Timmermans and Ross, Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries, Trends in Biotechnology (2019), https://doi.org/10.1016/j.tibtech.2019.11.003
Trends in Biotechnology
Figure 1. Workflow for Generating and Testing Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/CRISPR-Associated Protein 9 (Cas9) Knockout Actinobacterial Mutants for Antibiotic Discovery.
generates a mutant with a metabolic ‘clean slate’, simplifying the purification and characterization of less abundant molecules. Notably, one streptothricin-deletion mutant retained partial antibiotic activity, did not appear to produce any known antibiotics, and displayed a drastically different secondary metabolome compared with the wildtype strain. While Culp and colleagues have not yet described characterization of a novel antibiotic from this mutant, they appear to have found one. Metabolic changes in a mutant strain arising from BGC inactivation could be hypothesized to alter metabolic flux within the cell so that carbon may be funneled to other biosynthetic pathways. For example, by halting production of the aminoglycoside streptomycin, it is conceivable that the sugarbased precursors used in its biosynthesis could be catabolized and reenter central metabolism; the carbon could then be used for synthesis of the polyketide thiolactomycin, which is derived from acetate, methylmalonate, and cysteine [9]. Similar metabolic logic could be applied to streptothricin, which is likely derived from arginine, lysine, and glucosamine [10], and abolition of its biosynthesis correlates with an upregulation in the production of two other aminoglycosides, amicetin 2
and bamicetin, and the lysine-derived ferrioxamines. However, is it possible to rationally predict which classes of secondary metabolite may be upregulated by a certain type of metabolic knockout? Could genome mining and metabolic modeling be used in concert with strategic knockouts to prioritize strains that are more likely to show an altered secondary metabolite profile? Furthermore, is it likely that knocking out production of one antibiotic will result in the production of antibiotics with similar or different molecular targets? From an evolutionary standpoint, bacteria produce these compounds to compete with other microorganisms in their environment. However, do the encoded silent gene clusters have redundant antibiotic mechanisms of action? Or might they produce antibiotics with complementary bioactivities depending on the microbial competitors they interact with? In light of these questions, it is important to test CRISPR deletion strains generated for the activation of new natural products against a panel of pathogenic (and potentially drug-resistant) microbes to ensure that newly uncovered antibiotics are not missed by only screening for activity against the original antibiotic target.
Trends in Biotechnology, -- 2019, Vol. --, No. --
Nothing is more exciting to scientists than unanswered questions. Insights such as those provided here suggest that many long-abandoned environmental microbes languishing in libraries may have a second chance to reveal their true medicinal potential. 1Department
of Chemistry, Queen’s University, Kingston, ONT, Canada
@Twitter:
@avena.ross
*Correspondence: [email protected] https://doi.org/10.1016/j.tibtech.2019.11.003 ª 2019 Elsevier Ltd. All rights reserved.
References 1. Culp, E.J. et al. (2019) Hidden antibiotics in actinomycetes can be identified by inactivation of gene clusters for common antibiotics. Nat. Biotechnol. 37, 1149–1154 2. Blin, K. et al. (2019) The antiSMASH database version 2: a comprehensive resource on secondary metabolite biosynthetic gene clusters. Nucleic Acids Res. 47, D625–D630 3. Tomm, H.A. et al. (2019) Advances in microbial culturing conditions to activate silent biosynthetic gene clusters for novel metabolite production. J. Ind. Microbiol. Biotechnol. 46, 1381–1400 4. Mao, D. et al. (2018) Recent advances in activating silent biosynthetic gene clusters in bacteria. Curr. Opin. Microbiol. 45, 156–163 5. Wang, B. et al. (2019) Activation of silent biosynthetic gene clusters using transcription factor decoys. Nat. Chem. Biol. 15, 111–114 6. Lee, N. et al. (2019) Synthetic biology tools for novel secondary metabolite discovery in Streptomyces. J. Microbiol. Biotechnol. 29, 667–686
Please cite this article in press as: Timmermans and Ross, Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries, Trends in Biotechnology (2019), https://doi.org/10.1016/j.tibtech.2019.11.003
Trends in Biotechnology
7. Baltz, R.H. (2006) Marcel Faber Roundtable: is our antibiotic pipeline unproductive because of starvation, constipation or lack of inspiration? J. Ind. Microbiol. Biotechnol. 33, 507–513 8. Wang, M. et al. (2016) Sharing and community curation of mass spectrometry
data with Global Natural Products Social Molecular Networking. Nat. Biotechnol. 34, 828–837 9. Brown, M.S. et al. (2003) Biosynthetic origins of the natural product, thiolactomycin: a unique and selective inhibitor of type II dissociated fatty acid
synthases. J. Am. Chem. Soc. 125, 10166– 10167 10. Li, J. et al. (2013) Mining of a streptothricin gene cluster from Streptomyces sp. TP-A0356 genome via heterologous expression. Sci. China Life Sci. 56, 619–627
Trends in Biotechnology, -- 2019, Vol. --, No. --
3
Trends in Biotechnology
Spotlight
Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries Marshall L. Timmermans1 and Avena C. Ross1,@,* The redundancy of natural product biosynthesis in microbes poses a practical challenge for discovering new antimicrobial compounds from bacteria. The recent application of clustered regularly interspaced short palindromic repeats (CRISPR) technology by Culp et al. to inactivate the production of abundant antibiotics generates a metabolic clean slate for the detection and/ or isolation of new and less plentiful antibiotics activated in mutant strains.
Bacteria are the main source of antibiotics used in the clinic today. However, conventional natural product drug discovery pipelines have been abandoned and, coupled with the increase in bacterial pathogens resistant to commonly used antibiotics, this has generated a profound challenge for modern medicine. A CRISPR-Cas9 approach [1] now shows that interrupting production of highly abundant natural antibiotic products produced by actinomycetes stimulates the biosynthesis of other natural products, potentially reinvigorating antibiotic discovery from strains discarded from conventional drug discovery pipelines. While one approach for identifying novel bioactive natural products from bacteria is to investigate new and exotic bacteria, recent insights from genome sequencing indicate that it is also worthwhile reinvestigating the untapped biosynthetic potential of previously characterized strains. Analysis
shows that actinobacterial genomes encode, on average, 13 natural product biosynthetic gene clusters (BGCs), although they might only produce one or two natural products during laboratory fermentation [2]. Stimulating the biosynthesis of new natural products arising from these so-called ‘silent’ BGCs through various physiological and genetic means has been a high priority for the research community in recent years. Typically, this is done by exposing the bacteria to simulated environmental stressors, such as extremes of temperature, pH, toxic metal ions, or foreign bacterial species [3]. Alternatively, genetic manipulation of bacterial strains is used to alter the regulation of silent BGCs. This can be done by inserting constitutive promoters ahead of uncharacterized BGCs [4], engineering of transcription factors and ribosomes to alter global transcription and translation patterns [5], or by heterologous expression of either intact or refactored BGCs [6]. Culp and colleagues recently described a genetic strategy to increase the production of new actinobacterial natural products by inactivating the biosynthetic machinery for otherwise abundant confounding characterized antibiotics (Figure 1) [1]. Several of the most commonly rediscovered antibiotics in traditional drug discovery screens, the aminoglycosides streptomycin and streptothricin [7], were chosen because they are produced by 1% and 10% of actinomycetes, respectively. Using CRISPR Cas9 methodology, targeted disruption of genes in the streptomycin or streptothricin BGCs in a series of actinomycetes capable of producing the respective antibiotics was achieved. By using highly conserved sequences within the genes of the streptomycin and streptothricin BGCs, a single set of genetic constructs can be used for high-throughput deletion of related
pathways in many strains. This approach will enable the expedited drug discovery screens of large libraries in the future without requiring individually tailored CRISPR constructs or genome sequences. Mutant and wild-type strains can then be subjected to antibiotic activity assays and chemical analysis using the Global Natural Products Social Molecular Networking pipeline (GNPS) to identify newly produced or upregulated secondary metabolites/antibiotics [8]. All knockout strains generated produce different compound profiles compared with wild-type, but, interestingly there is also variation depending on which part of the antibiotic BGC is eliminated. This supports the hypothesis that abolishing the production of abundant secondary metabolites alters the flux of metabolic precursors within the cell and/or the genetic regulatory circuits responsive to these metabolites. The success of any discovery platform is measured in the natural products identified therein. Culp and colleagues identified the new production of known compounds from their CRISPR mutants, including the thiotetronate antibiotic thiolactomycin, the halogenated amino acid derivative 5-chloro3-formyl indole, and the aglycone of the cytotoxic jadomycin derivative phenanthroviridin. Additional upregulation in the production of the rare antibiotics bamicetin and amicetin, and numerous variants of the siderophore ferrrioxamine were also observed. While the structures and antibiotic activity of these particular compounds are already well characterized, these recent results suggest that the inactivation approach is effective at both activating the expression and increasing the titer of minor natural products [1]. Additionally, abolishing the production of an abundant natural product also
Trends in Biotechnology, -- 2019, Vol. --, No. --
1
Please cite this article in press as: Timmermans and Ross, Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries, Trends in Biotechnology (2019), https://doi.org/10.1016/j.tibtech.2019.11.003
Trends in Biotechnology
Figure 1. Workflow for Generating and Testing Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/CRISPR-Associated Protein 9 (Cas9) Knockout Actinobacterial Mutants for Antibiotic Discovery.
generates a mutant with a metabolic ‘clean slate’, simplifying the purification and characterization of less abundant molecules. Notably, one streptothricin-deletion mutant retained partial antibiotic activity, did not appear to produce any known antibiotics, and displayed a drastically different secondary metabolome compared with the wildtype strain. While Culp and colleagues have not yet described characterization of a novel antibiotic from this mutant, they appear to have found one. Metabolic changes in a mutant strain arising from BGC inactivation could be hypothesized to alter metabolic flux within the cell so that carbon may be funneled to other biosynthetic pathways. For example, by halting production of the aminoglycoside streptomycin, it is conceivable that the sugarbased precursors used in its biosynthesis could be catabolized and reenter central metabolism; the carbon could then be used for synthesis of the polyketide thiolactomycin, which is derived from acetate, methylmalonate, and cysteine [9]. Similar metabolic logic could be applied to streptothricin, which is likely derived from arginine, lysine, and glucosamine [10], and abolition of its biosynthesis correlates with an upregulation in the production of two other aminoglycosides, amicetin 2
and bamicetin, and the lysine-derived ferrioxamines. However, is it possible to rationally predict which classes of secondary metabolite may be upregulated by a certain type of metabolic knockout? Could genome mining and metabolic modeling be used in concert with strategic knockouts to prioritize strains that are more likely to show an altered secondary metabolite profile? Furthermore, is it likely that knocking out production of one antibiotic will result in the production of antibiotics with similar or different molecular targets? From an evolutionary standpoint, bacteria produce these compounds to compete with other microorganisms in their environment. However, do the encoded silent gene clusters have redundant antibiotic mechanisms of action? Or might they produce antibiotics with complementary bioactivities depending on the microbial competitors they interact with? In light of these questions, it is important to test CRISPR deletion strains generated for the activation of new natural products against a panel of pathogenic (and potentially drug-resistant) microbes to ensure that newly uncovered antibiotics are not missed by only screening for activity against the original antibiotic target.
Trends in Biotechnology, -- 2019, Vol. --, No. --
Nothing is more exciting to scientists than unanswered questions. Insights such as those provided here suggest that many long-abandoned environmental microbes languishing in libraries may have a second chance to reveal their true medicinal potential. 1Department
of Chemistry, Queen’s University, Kingston, ONT, Canada
@Twitter:
@avena.ross
*Correspondence: [email protected] https://doi.org/10.1016/j.tibtech.2019.11.003 ª 2019 Elsevier Ltd. All rights reserved.
References 1. Culp, E.J. et al. (2019) Hidden antibiotics in actinomycetes can be identified by inactivation of gene clusters for common antibiotics. Nat. Biotechnol. 37, 1149–1154 2. Blin, K. et al. (2019) The antiSMASH database version 2: a comprehensive resource on secondary metabolite biosynthetic gene clusters. Nucleic Acids Res. 47, D625–D630 3. Tomm, H.A. et al. (2019) Advances in microbial culturing conditions to activate silent biosynthetic gene clusters for novel metabolite production. J. Ind. Microbiol. Biotechnol. 46, 1381–1400 4. Mao, D. et al. (2018) Recent advances in activating silent biosynthetic gene clusters in bacteria. Curr. Opin. Microbiol. 45, 156–163 5. Wang, B. et al. (2019) Activation of silent biosynthetic gene clusters using transcription factor decoys. Nat. Chem. Biol. 15, 111–114 6. Lee, N. et al. (2019) Synthetic biology tools for novel secondary metabolite discovery in Streptomyces. J. Microbiol. Biotechnol. 29, 667–686
Please cite this article in press as: Timmermans and Ross, Unearthing Hidden Chemical Potential from Discarded Actinobacterial Libraries, Trends in Biotechnology (2019), https://doi.org/10.1016/j.tibtech.2019.11.003
Trends in Biotechnology
7. Baltz, R.H. (2006) Marcel Faber Roundtable: is our antibiotic pipeline unproductive because of starvation, constipation or lack of inspiration? J. Ind. Microbiol. Biotechnol. 33, 507–513 8. Wang, M. et al. (2016) Sharing and community curation of mass spectrometry
data with Global Natural Products Social Molecular Networking. Nat. Biotechnol. 34, 828–837 9. Brown, M.S. et al. (2003) Biosynthetic origins of the natural product, thiolactomycin: a unique and selective inhibitor of type II dissociated fatty acid
synthases. J. Am. Chem. Soc. 125, 10166– 10167 10. Li, J. et al. (2013) Mining of a streptothricin gene cluster from Streptomyces sp. TP-A0356 genome via heterologous expression. Sci. China Life Sci. 56, 619–627
Trends in Biotechnology, -- 2019, Vol. --, No. --
3