INSIGHTS | P E R S P E C T I V E S
MICROBIOLOGY
O Rh
P P
C
Rh
P
HO
O
Ph
R’
5
H P
6
R’
H
R’
H
O
Persisters unmasked
O H
P
Rh 4
O
Ph
O
O H
P P 7
O
Transfer hydroformylation
H
R
R’
Rh
P
C
P
O
HO Rh
H Ph
O R
3
R’
H P
R
P
H
P
C
10
C
Rh
P 2
O
H
O
O
P
P
R H
H2 (gas)
P
Rh
C
CO (gas)
O
8
H
R
Hydroformylation
H
Rh
P
C
1
P
R
Rh
O
P
C Rh
C O R
O H
O
R
9 H
Gas versus liquid phase. Hydroformylation to generate aldehydes uses gas-phase reactants, where transfer hydroformylation developed by Murphy et al. proceeds in solution.
dant but complex aldehyde in the presence of an inexpensive donor alkene to yield a complex and valuable product alkene. For example, Murphy et al. demonstrate a threestep transformation of (+)-yohimbine to (+)-yohimbenone that features dehydroformylation with norbornene as the acceptor. Alternatively, a simple alkene may be converted into a more valuable aldehyde by transfer from an abundant sacrificial aldehyde. Ultimately, this application is the more desirable because it would replace hydroformylation with a gasless equivalent. However, many questions must be addressed before transfer hydroformylation with inexpensive aldehydes provides a valuable alternative to hydroformylation under pressurized conditions. Can transfer hydroformylation proceed with bulky alkene acceptors? Can simple aldehydes such as propanal, or even formaldehyde or sugars, be used as aldehyde donors? Is useful con30
trol of regio- and enantioselectivity possible with different rhodium ligands, with acid promoters, or both? These new catalysts may provide the previously missing starting point for the development of general and completely gasless transfer hydroformylation processes. ■ REFERENCES AND NOTES
1. R. Franke, D. Selent, A. Börner, Chem. Rev. 112, 5675 (2012). 2. S. K. Murphy, J.-W. Park, F. A. Cruz, V. M. Dong, Science 347, 56 (2015). 3. A. E. Braunstein, M. G. Kritzman, Enzymologia 2, 129 (1937). 4. S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 117, 7562 (1995). 5. M. C. Willis, Chem. Rev. 110, 725 (2010). 6. Dehydroformylation is thermodynamically uphill. Conversion of propanal to ethylene, CO, and H2 has standard enthalpy and free energy of +31.2 and +13.9 kcal/mol, respectively. Data from the NIST Gas Phase Thermochemistry Database (7). 7. http://webbook.nist.gov/chemistry/guide/#thermo-gas 10.1126/science.aaa2329
Intracellular toxins cause bacterial growth arrest and antibiotic tolerance By David W. Holden
P
enicillin may have saved more human lives than any other drug. Yet, almost as soon as it was introduced in the 1940s, researchers found that the antibiotic could not completely sterilize a culture of a Staphylococcus aureus strain sensitive to the drug (1). Shortly thereafter, Joseph Bigger showed that when the few cells that had survived an initial treatment were regrown in the absence of penicillin and then exposed again to the antibiotic, the proportion of survivors was similar to that found after the first treatment (see the first figure). Therefore, the survivors were not stable drug-resistant mutants, but transient drugtolerant persisters (2). In the past decade, a resurgence of interest in persisters has revealed some of the molecular mechanisms that stimulate their formation. It has become clear that intracellular toxins present in virtually all bacteria control reversible bacterial growth arrest, explaining their antibiotic tolerance. In genetically similar or identical populations of organisms that have the same phenotype, an infection or other stress that has the potential to kill one organism can in theory kill the entire population. The consequences of this frailty are exemplified by the vast monocultures of wheat devastated by fungal disease epidemics in the North American Great Plains in the early 20th century. Bacteria reproduce clonally and might therefore appear to be similarly vulnerable to attack from other bacteria or bacteriophages, sudden environmental changes, or exposure to antibiotics. These potentially lethal stresses are very likely to have imposed immense selective pressure for the evolution of phenotypic heterogeneity among bacteria. Since Bigger’s experiment, numerous studies have found cell-to-cell variation involving different physiological processes in clonal populations of bacteria grown in the same environment (3, 4). Virtually all bacteria form antibiotictolerant persisters. Many aspects of their sciencemag.org SCIENCE
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
Downloaded from www.sciencemag.org on January 1, 2015
P
Antibiotic
No antibiotic
Antibiotic
Stress Replicating cells
Nonreplicating persisters
Antibiotic-resistant mutants
ILLUSTRATIONS: P. HUEY/SCIENCE
Resistance and persisters. Bacterial populations can contain antibiotic-resistant mutants and/or nonreplicating persisters among replicating cells. Antibiotics kill replicating antibiotic-sensitive bacteria but not resistant mutants or persisters. Resistance is stably inherited among bacteria in the presence or absence of antibiotic, whereas persisters resume growth in the absence of antibiotic. Unstressed populations can continue to generate persisters at low frequency, while various stresses stimulate persister frequency.
Section of Microbiology, MRC Centre for Molecular Bacteriology and Infection, Imperial College London, Armstrong Road, London SW7 2AZ, UK. E-mail: d.holden@ imperial.ac.uk
part of the bacterial population (9). In addition, nutritional starvation and possibly other stresses stimulate persister formation by activating a pathway involving the intracellular signaling molecule ppGpp, polyphosphate, and the cellular protease Lon that degrades antitoxins (10). A process called conditional cooperativity exerts exquisite control over the autoregulated expression of TA genes. For example, at low concentrations, the Doc toxin binds to its antitoxin (Phd) to form an efficient transcriptional repressor complex on the phd-doc operator. However, at higher concentrations, more Doc molecules bind to the antitoxin such that it can no longer bind DNA, thereby derepressing transcription (11). Thus, the toxin:antitoxin ratio fine-tunes transcription in response to changing levels of unbound toxin (see the second figure). Presumably, exit from the growth-arrested state is initiated by an increase in antitoxin levels and resultant inactivation of the toxin, but this has not yet been shown. Increasing toxin concentration
complex biology have now been revealed, especially with the laboratory workhorse, Escherichia coli. These advances have been facilitated by equipment and methods that enable single-cell analysis of bacteria. Perhaps unsurprisingly, most persisters are slow- or nongrowing cells (5). Nongrowing cells tend to tolerate a wide variety of stresses. Their production in replicating populations of bacteria is unlikely to have evolved in response to selective pressure from antibiotics. But a lack of growth usually involves reduced metabolic activity, and metabolically inactive cells have greater tolerance to antibiotics than replicating cells. Persisters are present in low numbers in apparently nonstressed populations, before exposure to antibiotics (5). This can be viewed as a bacterial insurance against the arrival of a potentially lethal stress. Entry into the persister state is often controlled by gene pairs encoding cognate toxin-antitoxins (TAs) (6, 7). During bacterial replication, protein toxins are mainly bound to their protein or RNA antitoxins, preventing toxin activity. When liberated, the toxins variously inhibit DNA replication, cleave mRNA endoribonucleases, block protein translation, or interfere with the cell cytoskeleton (8) (see the second figure). The common physiological consequence of these activities is a transient arrest of cell growth. Studies on the protein-based TA systems suggest that inherent instability of an antitoxin leads to random fluctuation in the amounts of free and bound toxin just below a critical threshold. In rare cases, the threshold is surpassed, generating sufficient free toxin to cause growth arrest in a very small
The near-ubiquitous occurrence of TA genes in bacteria and the numerous variations of these genes harbored by many bacteria show clearly that they are both ancient and of fundamental importance to bacteria. They are also likely to have important functions during infection, when bacterial pathogens encounter highly stressful environments in their hosts. Some strains of Mycobacterium tuberculosis, which is notorious for causing chronic and recurrent infections, contain up to 79 such genes, but detailed investigations into the functions of these and TA genes of other pathogens have only begun recently. Salmonella Typhi and Salmonella Typhimurium also cause recurrent infections in humans. Whole-genome sequencing of S. Typhimurium isolates from patients has provided unambiguous evidence for reappearance of the original strain after repeated antibiotic treatment, directly implicating persisters (12). Use of fluorescence-based techniques with the mouse model of typhoid-like disease caused by S. Typhimurium has enabled direct observation and characterization of persisters during infection (13, 14). Salmonella has adapted to survive and replicate in vacuoles after phagocytosis by macrophages. The low pH and poor nutritional status of the vacuole greatly enhance the frequency of TA-dependent nonreplicating persisters (14). Reporters of metabolic activity suggest a range of physiological states among nonreplicating intracellular Salmonella, from persisters primed for immediate resumption of growth in new macrophages, to cells that are metabolically inactive (dormant) and could require specific resuscitation factors (14). Bigger’s insights into persisters were remarkable. Using rudimentary microbio-
DNA
Active toxin
Inactive toxin Antitoxin
Stress Degraded antitoxin mRNA
Dual functions of toxins and antitoxins. During normal cell growth, antitoxin and toxin are expressed from a DNA operon. Toxin activity is neutralized by binding antitoxin. Free antitoxin is a weak autotranscriptional repressor. Transcriptional repression is enhanced by increasing concentration of cognate toxin (which forms a complex with the antitoxin on the operator DNA). Above a toxin concentration threshold, toxin-antitoxin complexes dissociate from DNA, derepressing transcription. Thus, cooperative binding of toxin-antitoxin to DNA depends on toxin concentration. Antitoxin is degraded in response to stress, liberating activated toxin. This arrests cell growth—for example, by cleaving mRNA.
SCIENCE sciencemag.org
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
31
INSIGHTS | P E R S P E C T I V E S
logical techniques, he distinguished persisters from resistant mutants, showed that their levels could be enhanced by stress, and anticipated that they would be in a “dormant nondividing phase” and present among other bacterial pathogens (2). The recent exciting progress on mechanisms of TA function establishes these toxins as key inducers of the persister state. Future research should elucidate the many functions of TAs and how they work collectively during persistent bacterial infections. For example, it is unclear whether different stresses activate different TA subsets, and what the profiles of toxin activation are in individual bacterial cells. Some toxins have been shown to be sequence-specific ribonucleases, but whether this specificity has physiological implications is uncertain. It could be that bacteria perceive signals that trigger their exit from quiescence, but the mechanisms involved are unknown. If persisters lead to recurrent infections requiring multiple courses of antibiotics, then they are likely to contribute appreciably to the current worldwide crisis of antibiotic resistance. Yet, surprisingly little is known about the relative usage of antibiotics for persistent infections and the degree to which persisters influence the emergence of resistance. In the long term, TAs and associated signaling molecules may provide targets for drugs that can either prevent persisters from being formed, or—perhaps more feasibly—coax them out of the nonreplicating state so that they resume susceptibility to antibiotics. This might finally enable complete eradication of an otherwise recurrent or persistent infection, so that, as Bigger put it, “the success of penicillin therapy will become more commensurate with its potentialities” (2). ■ REFERENCES AND NOTES
1. G. L. Hobby, K. Meyer, E. Chaffee, Exp. Biol. Med. 50, 281 (1942). 2. J. W. Bigger, Lancet 244, 497 (1944). 3. J. L. Spudich, D. E. Koshland Jr., Nature 262, 467 (1976). 4. J. Casadesús, D. A. Low, J. Biol. Chem. 288, 13929 (2013). 5. N. Q. Balaban, J. Merrin, R. Chait, L. Kowalik, S. Leibler, Science 305, 1622 (2004). 6. H. S. Moyed, K. P. Bertrand, J. Bacteriol. 155, 768 (1983). 7. E. Maisonneuve, L. J. Shakespeare, M. G. Jørgensen, K. Gerdes, Proc. Natl. Acad. Sci. U.S.A. 108, 13206 (2011). 8. E. Maisonneuve, K. Gerdes, Cell 157, 539 (2014). 9. E. Rotem et al., Proc. Natl. Acad. Sci. U.S.A. 107, 12541 (2010). 10. E. Maisonneuve, M. Castro-Camargo, K. Gerdes, Cell 154, 1140 (2013). 11. A. Garcia-Pino et al., Cell 142, 101 (2010). 12. C. K. Okoro et al., Clin. Infect. Dis. 54, 955 (2012). 13. B. Claudi et al., Cell 158, 722 (2014). 14. S. Helaine et al., Science 343, 204 (2014). ACKNOWLEDGMENTS
I thank S. Helaine and T. Thurston for helpful comments. 10.1126/science.1262033
32
CELL BIOLOGY
Lysosomal lipid lengthens life span A fatty acid moves from the lysosome to the nucleus, altering gene expression and extending longevity in the worm By Shuo Han and Anne Brunet
L
ysosomes were discovered more than 60 years ago as highly acidic cellular organelles containing many enzymes responsible for breaking down macromolecules (1). Since then, their roles have expanded. Lysosomes function in autophagy, the process that breaks down cellular components to allow cell survival and homeostasis in the face of starvation (1). These organelles also have emerged as a signaling hub for the enzyme mechanistic target of rapamycin (mTOR), a protein kinase involved in cellular and organismal growth responses to nutrient availability (2). We also now recognize links between aberrant lysosomal function and several diseases, including lysosomal storage diseases (e.g., Tay-Sachs disease) and neurodegenerative disorders (e.g., Parkinson’s disease), and also with aging (1). On page 83 of this issue, Folick et al. (3) indicate how lysosomes play a role in the latter—by deploying a lipid molecule to the nucleus, whose impact on gene expression extends life span in an animal model (the nematode Caenorhabditis elegans). The study not only uncovers a lysosome-to-nucleus signaling pathway but also highlights the potential of lipids in mediating long-range physiological effects. Lysosomes contain about 60 enzymes, including many well-conserved lipases involved in fatty acid breakdown. Defects in lysosomal acid lipase A (LIPA) lead to several human lysosomal storage diseases, including Wolman disease, a disorder characterized by metabolic defects and death in childhood (4). In C. elegans, the LIPA homolog LIPL-4 is highly expressed in specific conditions that are linked to life-span extension (5, 6). However, the mechanism by which this enzyme modulates aging has remained elusive. Using a combination of genetics, metabolomics, biochemistry, and immunocytochemistry, Folick et al. explored the molecular mechanisms by which lysosomal LIPL-4 activation regulates aging in C. elegans. They show that worms overexpressing LIPL-4 live substantially longer than normal worms and produce increased amounts of several bioactive lipids, notably the fatty acid oleoyletha-
nolamide (OEA). OEA is likely generated by the breakdown of more complex lipids in the lysosome by LIPL-4. LIPL-4–overexpressing worms also exhibit an increased amount of a fatty acid binding protein called lipidbinding protein-8 [(LBP-8); the human homolog is fatty acid binding protein (FABP)]. Elegantly coupling fluorescence imaging with mutations that alter protein targeting to the lysosome, Folick et al. demonstrate that LIPL-4 must reside within the lysosome to extend life span. By contrast, LBP-8 translocates from the lysosome into the nucleus to ensure increased longevity. As LBP-8 can directly bind to OEA, these results suggest that LBP-8 is a lipid chaperone assisting OEA entry into the nucleus (see the figure). What happens once OEA is shuttled into the nucleus? Folick et al. found that OEA
“…dietary modulation of fatty acids…has the potential to delay aging.” physically binds to and activates conserved nuclear hormone receptors, thereby activating the transcription of target genes. Fatty acid ligands have been reported to control the transcriptional activity of subfamilies of nuclear receptors (7), and OEA can bind to the nuclear receptor peroxisome proliferator–activated receptor-α (PPARα) in mice (8). The authors report that two particular nuclear receptors—nuclear hormone receptor-49 (NHR-49) and NHR-80, the C. elegans homologs of mammalian PPARα and hepatic nuclear factor 4, respectively—are both required for LIPL-4–induced longevity, and that OEA can directly bind to NHR-80. This observation is consistent with previous reports that NHR-49 and NHR-80 play critical roles in life-span regulation in C. elegans (9, 10). What about dietary supplementation of OEA? Folick et al. found that feeding worms OEA during their adult life is sufficient to Department of Genetics, Stanford University, Stanford, CA 94035, USA. E-mail: [email protected]
sciencemag.org SCIENCE
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
Persisters unmasked David W. Holden Science 347, 30 (2015); DOI: 10.1126/science.1262033
If you wish to distribute this article to others, you can order high-quality copies for your colleagues, clients, or customers by clicking here. Permission to republish or repurpose articles or portions of articles can be obtained by following the guidelines here. The following resources related to this article are available online at www.sciencemag.org (this information is current as of January 1, 2015 ): Updated information and services, including high-resolution figures, can be found in the online version of this article at: http://www.sciencemag.org/content/347/6217/30.full.html This article cites 14 articles, 8 of which can be accessed free: http://www.sciencemag.org/content/347/6217/30.full.html#ref-list-1 This article appears in the following subject collections: Microbiology http://www.sciencemag.org/cgi/collection/microbio
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2015 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.
Downloaded from www.sciencemag.org on January 1, 2015
This copy is for your personal, non-commercial use only.
MICROBIOLOGY
O Rh
P P
C
Rh
P
HO
O
Ph
R’
5
H P
6
R’
H
R’
H
O
Persisters unmasked
O H
P
Rh 4
O
Ph
O
O H
P P 7
O
Transfer hydroformylation
H
R
R’
Rh
P
C
P
O
HO Rh
H Ph
O R
3
R’
H P
R
P
H
P
C
10
C
Rh
P 2
O
H
O
O
P
P
R H
H2 (gas)
P
Rh
C
CO (gas)
O
8
H
R
Hydroformylation
H
Rh
P
C
1
P
R
Rh
O
P
C Rh
C O R
O H
O
R
9 H
Gas versus liquid phase. Hydroformylation to generate aldehydes uses gas-phase reactants, where transfer hydroformylation developed by Murphy et al. proceeds in solution.
dant but complex aldehyde in the presence of an inexpensive donor alkene to yield a complex and valuable product alkene. For example, Murphy et al. demonstrate a threestep transformation of (+)-yohimbine to (+)-yohimbenone that features dehydroformylation with norbornene as the acceptor. Alternatively, a simple alkene may be converted into a more valuable aldehyde by transfer from an abundant sacrificial aldehyde. Ultimately, this application is the more desirable because it would replace hydroformylation with a gasless equivalent. However, many questions must be addressed before transfer hydroformylation with inexpensive aldehydes provides a valuable alternative to hydroformylation under pressurized conditions. Can transfer hydroformylation proceed with bulky alkene acceptors? Can simple aldehydes such as propanal, or even formaldehyde or sugars, be used as aldehyde donors? Is useful con30
trol of regio- and enantioselectivity possible with different rhodium ligands, with acid promoters, or both? These new catalysts may provide the previously missing starting point for the development of general and completely gasless transfer hydroformylation processes. ■ REFERENCES AND NOTES
1. R. Franke, D. Selent, A. Börner, Chem. Rev. 112, 5675 (2012). 2. S. K. Murphy, J.-W. Park, F. A. Cruz, V. M. Dong, Science 347, 56 (2015). 3. A. E. Braunstein, M. G. Kritzman, Enzymologia 2, 129 (1937). 4. S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 117, 7562 (1995). 5. M. C. Willis, Chem. Rev. 110, 725 (2010). 6. Dehydroformylation is thermodynamically uphill. Conversion of propanal to ethylene, CO, and H2 has standard enthalpy and free energy of +31.2 and +13.9 kcal/mol, respectively. Data from the NIST Gas Phase Thermochemistry Database (7). 7. http://webbook.nist.gov/chemistry/guide/#thermo-gas 10.1126/science.aaa2329
Intracellular toxins cause bacterial growth arrest and antibiotic tolerance By David W. Holden
P
enicillin may have saved more human lives than any other drug. Yet, almost as soon as it was introduced in the 1940s, researchers found that the antibiotic could not completely sterilize a culture of a Staphylococcus aureus strain sensitive to the drug (1). Shortly thereafter, Joseph Bigger showed that when the few cells that had survived an initial treatment were regrown in the absence of penicillin and then exposed again to the antibiotic, the proportion of survivors was similar to that found after the first treatment (see the first figure). Therefore, the survivors were not stable drug-resistant mutants, but transient drugtolerant persisters (2). In the past decade, a resurgence of interest in persisters has revealed some of the molecular mechanisms that stimulate their formation. It has become clear that intracellular toxins present in virtually all bacteria control reversible bacterial growth arrest, explaining their antibiotic tolerance. In genetically similar or identical populations of organisms that have the same phenotype, an infection or other stress that has the potential to kill one organism can in theory kill the entire population. The consequences of this frailty are exemplified by the vast monocultures of wheat devastated by fungal disease epidemics in the North American Great Plains in the early 20th century. Bacteria reproduce clonally and might therefore appear to be similarly vulnerable to attack from other bacteria or bacteriophages, sudden environmental changes, or exposure to antibiotics. These potentially lethal stresses are very likely to have imposed immense selective pressure for the evolution of phenotypic heterogeneity among bacteria. Since Bigger’s experiment, numerous studies have found cell-to-cell variation involving different physiological processes in clonal populations of bacteria grown in the same environment (3, 4). Virtually all bacteria form antibiotictolerant persisters. Many aspects of their sciencemag.org SCIENCE
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
Downloaded from www.sciencemag.org on January 1, 2015
P
Antibiotic
No antibiotic
Antibiotic
Stress Replicating cells
Nonreplicating persisters
Antibiotic-resistant mutants
ILLUSTRATIONS: P. HUEY/SCIENCE
Resistance and persisters. Bacterial populations can contain antibiotic-resistant mutants and/or nonreplicating persisters among replicating cells. Antibiotics kill replicating antibiotic-sensitive bacteria but not resistant mutants or persisters. Resistance is stably inherited among bacteria in the presence or absence of antibiotic, whereas persisters resume growth in the absence of antibiotic. Unstressed populations can continue to generate persisters at low frequency, while various stresses stimulate persister frequency.
Section of Microbiology, MRC Centre for Molecular Bacteriology and Infection, Imperial College London, Armstrong Road, London SW7 2AZ, UK. E-mail: d.holden@ imperial.ac.uk
part of the bacterial population (9). In addition, nutritional starvation and possibly other stresses stimulate persister formation by activating a pathway involving the intracellular signaling molecule ppGpp, polyphosphate, and the cellular protease Lon that degrades antitoxins (10). A process called conditional cooperativity exerts exquisite control over the autoregulated expression of TA genes. For example, at low concentrations, the Doc toxin binds to its antitoxin (Phd) to form an efficient transcriptional repressor complex on the phd-doc operator. However, at higher concentrations, more Doc molecules bind to the antitoxin such that it can no longer bind DNA, thereby derepressing transcription (11). Thus, the toxin:antitoxin ratio fine-tunes transcription in response to changing levels of unbound toxin (see the second figure). Presumably, exit from the growth-arrested state is initiated by an increase in antitoxin levels and resultant inactivation of the toxin, but this has not yet been shown. Increasing toxin concentration
complex biology have now been revealed, especially with the laboratory workhorse, Escherichia coli. These advances have been facilitated by equipment and methods that enable single-cell analysis of bacteria. Perhaps unsurprisingly, most persisters are slow- or nongrowing cells (5). Nongrowing cells tend to tolerate a wide variety of stresses. Their production in replicating populations of bacteria is unlikely to have evolved in response to selective pressure from antibiotics. But a lack of growth usually involves reduced metabolic activity, and metabolically inactive cells have greater tolerance to antibiotics than replicating cells. Persisters are present in low numbers in apparently nonstressed populations, before exposure to antibiotics (5). This can be viewed as a bacterial insurance against the arrival of a potentially lethal stress. Entry into the persister state is often controlled by gene pairs encoding cognate toxin-antitoxins (TAs) (6, 7). During bacterial replication, protein toxins are mainly bound to their protein or RNA antitoxins, preventing toxin activity. When liberated, the toxins variously inhibit DNA replication, cleave mRNA endoribonucleases, block protein translation, or interfere with the cell cytoskeleton (8) (see the second figure). The common physiological consequence of these activities is a transient arrest of cell growth. Studies on the protein-based TA systems suggest that inherent instability of an antitoxin leads to random fluctuation in the amounts of free and bound toxin just below a critical threshold. In rare cases, the threshold is surpassed, generating sufficient free toxin to cause growth arrest in a very small
The near-ubiquitous occurrence of TA genes in bacteria and the numerous variations of these genes harbored by many bacteria show clearly that they are both ancient and of fundamental importance to bacteria. They are also likely to have important functions during infection, when bacterial pathogens encounter highly stressful environments in their hosts. Some strains of Mycobacterium tuberculosis, which is notorious for causing chronic and recurrent infections, contain up to 79 such genes, but detailed investigations into the functions of these and TA genes of other pathogens have only begun recently. Salmonella Typhi and Salmonella Typhimurium also cause recurrent infections in humans. Whole-genome sequencing of S. Typhimurium isolates from patients has provided unambiguous evidence for reappearance of the original strain after repeated antibiotic treatment, directly implicating persisters (12). Use of fluorescence-based techniques with the mouse model of typhoid-like disease caused by S. Typhimurium has enabled direct observation and characterization of persisters during infection (13, 14). Salmonella has adapted to survive and replicate in vacuoles after phagocytosis by macrophages. The low pH and poor nutritional status of the vacuole greatly enhance the frequency of TA-dependent nonreplicating persisters (14). Reporters of metabolic activity suggest a range of physiological states among nonreplicating intracellular Salmonella, from persisters primed for immediate resumption of growth in new macrophages, to cells that are metabolically inactive (dormant) and could require specific resuscitation factors (14). Bigger’s insights into persisters were remarkable. Using rudimentary microbio-
DNA
Active toxin
Inactive toxin Antitoxin
Stress Degraded antitoxin mRNA
Dual functions of toxins and antitoxins. During normal cell growth, antitoxin and toxin are expressed from a DNA operon. Toxin activity is neutralized by binding antitoxin. Free antitoxin is a weak autotranscriptional repressor. Transcriptional repression is enhanced by increasing concentration of cognate toxin (which forms a complex with the antitoxin on the operator DNA). Above a toxin concentration threshold, toxin-antitoxin complexes dissociate from DNA, derepressing transcription. Thus, cooperative binding of toxin-antitoxin to DNA depends on toxin concentration. Antitoxin is degraded in response to stress, liberating activated toxin. This arrests cell growth—for example, by cleaving mRNA.
SCIENCE sciencemag.org
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
31
INSIGHTS | P E R S P E C T I V E S
logical techniques, he distinguished persisters from resistant mutants, showed that their levels could be enhanced by stress, and anticipated that they would be in a “dormant nondividing phase” and present among other bacterial pathogens (2). The recent exciting progress on mechanisms of TA function establishes these toxins as key inducers of the persister state. Future research should elucidate the many functions of TAs and how they work collectively during persistent bacterial infections. For example, it is unclear whether different stresses activate different TA subsets, and what the profiles of toxin activation are in individual bacterial cells. Some toxins have been shown to be sequence-specific ribonucleases, but whether this specificity has physiological implications is uncertain. It could be that bacteria perceive signals that trigger their exit from quiescence, but the mechanisms involved are unknown. If persisters lead to recurrent infections requiring multiple courses of antibiotics, then they are likely to contribute appreciably to the current worldwide crisis of antibiotic resistance. Yet, surprisingly little is known about the relative usage of antibiotics for persistent infections and the degree to which persisters influence the emergence of resistance. In the long term, TAs and associated signaling molecules may provide targets for drugs that can either prevent persisters from being formed, or—perhaps more feasibly—coax them out of the nonreplicating state so that they resume susceptibility to antibiotics. This might finally enable complete eradication of an otherwise recurrent or persistent infection, so that, as Bigger put it, “the success of penicillin therapy will become more commensurate with its potentialities” (2). ■ REFERENCES AND NOTES
1. G. L. Hobby, K. Meyer, E. Chaffee, Exp. Biol. Med. 50, 281 (1942). 2. J. W. Bigger, Lancet 244, 497 (1944). 3. J. L. Spudich, D. E. Koshland Jr., Nature 262, 467 (1976). 4. J. Casadesús, D. A. Low, J. Biol. Chem. 288, 13929 (2013). 5. N. Q. Balaban, J. Merrin, R. Chait, L. Kowalik, S. Leibler, Science 305, 1622 (2004). 6. H. S. Moyed, K. P. Bertrand, J. Bacteriol. 155, 768 (1983). 7. E. Maisonneuve, L. J. Shakespeare, M. G. Jørgensen, K. Gerdes, Proc. Natl. Acad. Sci. U.S.A. 108, 13206 (2011). 8. E. Maisonneuve, K. Gerdes, Cell 157, 539 (2014). 9. E. Rotem et al., Proc. Natl. Acad. Sci. U.S.A. 107, 12541 (2010). 10. E. Maisonneuve, M. Castro-Camargo, K. Gerdes, Cell 154, 1140 (2013). 11. A. Garcia-Pino et al., Cell 142, 101 (2010). 12. C. K. Okoro et al., Clin. Infect. Dis. 54, 955 (2012). 13. B. Claudi et al., Cell 158, 722 (2014). 14. S. Helaine et al., Science 343, 204 (2014). ACKNOWLEDGMENTS
I thank S. Helaine and T. Thurston for helpful comments. 10.1126/science.1262033
32
CELL BIOLOGY
Lysosomal lipid lengthens life span A fatty acid moves from the lysosome to the nucleus, altering gene expression and extending longevity in the worm By Shuo Han and Anne Brunet
L
ysosomes were discovered more than 60 years ago as highly acidic cellular organelles containing many enzymes responsible for breaking down macromolecules (1). Since then, their roles have expanded. Lysosomes function in autophagy, the process that breaks down cellular components to allow cell survival and homeostasis in the face of starvation (1). These organelles also have emerged as a signaling hub for the enzyme mechanistic target of rapamycin (mTOR), a protein kinase involved in cellular and organismal growth responses to nutrient availability (2). We also now recognize links between aberrant lysosomal function and several diseases, including lysosomal storage diseases (e.g., Tay-Sachs disease) and neurodegenerative disorders (e.g., Parkinson’s disease), and also with aging (1). On page 83 of this issue, Folick et al. (3) indicate how lysosomes play a role in the latter—by deploying a lipid molecule to the nucleus, whose impact on gene expression extends life span in an animal model (the nematode Caenorhabditis elegans). The study not only uncovers a lysosome-to-nucleus signaling pathway but also highlights the potential of lipids in mediating long-range physiological effects. Lysosomes contain about 60 enzymes, including many well-conserved lipases involved in fatty acid breakdown. Defects in lysosomal acid lipase A (LIPA) lead to several human lysosomal storage diseases, including Wolman disease, a disorder characterized by metabolic defects and death in childhood (4). In C. elegans, the LIPA homolog LIPL-4 is highly expressed in specific conditions that are linked to life-span extension (5, 6). However, the mechanism by which this enzyme modulates aging has remained elusive. Using a combination of genetics, metabolomics, biochemistry, and immunocytochemistry, Folick et al. explored the molecular mechanisms by which lysosomal LIPL-4 activation regulates aging in C. elegans. They show that worms overexpressing LIPL-4 live substantially longer than normal worms and produce increased amounts of several bioactive lipids, notably the fatty acid oleoyletha-
nolamide (OEA). OEA is likely generated by the breakdown of more complex lipids in the lysosome by LIPL-4. LIPL-4–overexpressing worms also exhibit an increased amount of a fatty acid binding protein called lipidbinding protein-8 [(LBP-8); the human homolog is fatty acid binding protein (FABP)]. Elegantly coupling fluorescence imaging with mutations that alter protein targeting to the lysosome, Folick et al. demonstrate that LIPL-4 must reside within the lysosome to extend life span. By contrast, LBP-8 translocates from the lysosome into the nucleus to ensure increased longevity. As LBP-8 can directly bind to OEA, these results suggest that LBP-8 is a lipid chaperone assisting OEA entry into the nucleus (see the figure). What happens once OEA is shuttled into the nucleus? Folick et al. found that OEA
“…dietary modulation of fatty acids…has the potential to delay aging.” physically binds to and activates conserved nuclear hormone receptors, thereby activating the transcription of target genes. Fatty acid ligands have been reported to control the transcriptional activity of subfamilies of nuclear receptors (7), and OEA can bind to the nuclear receptor peroxisome proliferator–activated receptor-α (PPARα) in mice (8). The authors report that two particular nuclear receptors—nuclear hormone receptor-49 (NHR-49) and NHR-80, the C. elegans homologs of mammalian PPARα and hepatic nuclear factor 4, respectively—are both required for LIPL-4–induced longevity, and that OEA can directly bind to NHR-80. This observation is consistent with previous reports that NHR-49 and NHR-80 play critical roles in life-span regulation in C. elegans (9, 10). What about dietary supplementation of OEA? Folick et al. found that feeding worms OEA during their adult life is sufficient to Department of Genetics, Stanford University, Stanford, CA 94035, USA. E-mail: [email protected]
sciencemag.org SCIENCE
2 JANUARY 2015 • VOL 347 ISSUE 6217
Published by AAAS
Persisters unmasked David W. Holden Science 347, 30 (2015); DOI: 10.1126/science.1262033
If you wish to distribute this article to others, you can order high-quality copies for your colleagues, clients, or customers by clicking here. Permission to republish or repurpose articles or portions of articles can be obtained by following the guidelines here. The following resources related to this article are available online at www.sciencemag.org (this information is current as of January 1, 2015 ): Updated information and services, including high-resolution figures, can be found in the online version of this article at: http://www.sciencemag.org/content/347/6217/30.full.html This article cites 14 articles, 8 of which can be accessed free: http://www.sciencemag.org/content/347/6217/30.full.html#ref-list-1 This article appears in the following subject collections: Microbiology http://www.sciencemag.org/cgi/collection/microbio
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2015 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.
Downloaded from www.sciencemag.org on January 1, 2015
This copy is for your personal, non-commercial use only.
