- Email: [email protected]
Getting a Genomic View of DNA Replication Stress
Molecular Cell
Previews Getting a Genomic View of DNA Replication Stress Lee Zou1,2,* and Hai Dang Nguyen1 1Massachusetts
General Hospital Cancer Center, Harvard Medical School, Charlestown, MA 02129, USA of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02115 *Correspondence: [email protected] https://doi.org/10.1016/j.molcel.2018.10.001 2Department
DNA replication forks collapse at numerous sites throughout the genome under replication stress. Studies by Shastri et al. (2018) and Tubbs et al. (2018) used different genomics approaches to map the sites of replication fork collapse, revealing the contribution of specific DNA sequences to replication stress.
DNA replication stress is a major cause of genomic instability in the cell (Saldivar et al., 2017). Mounting evidence has shown that DNA replication stress arises from a variety of sources, such as dysregulation of replication initiation, defects in replication forks, compromised DNA repair or DNA damage signaling, and conflicts between replication and transcription. A number of chromosomal loci and DNA sequence elements have been implicated in the generation of replication stress and genomic instability (Glover et al., 2017; Neil et al., 2017). However, a genomic view of the major sites that impose replication stress is still lacking. Two independent studies by the laboratories of Brown and Nussenzweig have used different approaches to map the difficult-to-replicate sites in the genome, revealing a widespread contribution of specific DNA sequence elements to replication stress (Shastri et al., 2018; Tubbs et al., 2018). DNA replication forks encountering impediments on DNA commonly expose more single-stranded DNA (ssDNA), which is rapidly coated by the ssDNA-binding protein RPA. The RPA-coated ssDNA at stressed replication forks recruits the ATR-ATRIP kinase complex and promotes its activation, enabling ATR to stabilize stressed replication forks (Zou and Elledge, 2003). When ATR is inhibited, replication origin firing is increased and replication forks become prone to stalling and collapse at sites difficult to replicate, which may lead to RPA accumulation at these sites (Buisson et al., 2015; Casper et al., 2002). Using RPA accumulation as a surrogate for stalled/ collapsed replication forks, Shastri et al.
performed RPA chromatin immunoprecipitation sequencing (ChIP-seq) in mouse embryonic fibroblasts (MEFs) treated with ATR inhibitor and aphidicolin, an inhibitor of DNA polymerases, and mapped 171 ATR-dependent replication perturbed locations (RPLs) in the genome (Shastri et al., 2018) (Figure 1). Careful analysis of the RPLs revealed several common sequence features. Many RPLs contain microsatellite repeats near the center. These short tandem repeats are enriched for purines and specific DNA sequences, such as (CAGAGG/ and (CACAG/CTGTG)n CCTCTG)n (Figure 1). When analyzed in vitro, synthetic oligos of these repeats displayed features of secondary DNA structures and non-B-form DNA. Furthermore, the (CCTCTG)n repeats impeded the DNA synthesis by polymerase d in vitro and increased replication fork stalling on plasmids in cells. Collectively, these results suggest that the purine-rich microsatellite repeats in RPLs are natural barriers of replication forks. In addition to mapping RPLs with RPA ChIP-seq, Shastri et al. used another method called BrITL (breaks identified by TdT labeling) to identify the sites of DNA double-stranded breaks (DSBs) in MEFs treated with ATR inhibitor and aphidicolin (Shastri et al., 2018) (Figure 1). A subset of BrITL sites coincides with RPLs, suggesting that some of the DSBs induced by ATR inhibitor and aphidicolin arise from replication forks stalled at RPLs. However, BrITL also detected many DSBs that did not recruit RPA. These RPL-independent BrITL sites commonly contain long inverted repeats and quasi-palindromes with high melting
temperatures. Although these BrITL sites are distinct from RPLs in sequences, they share similarities in the repetitive nature and potential to form secondary DNA structures. Whether the RPA-free BrITL sites stall replication forks as RPLs still needs to be investigated. If replication forks were stalled at the BrITL sites, it would be important to understand why RPA is excluded from some of these sites and whether RPA exclusion contributes to DNA breakage. Furthermore, it would be interesting to investigate how ATR protects these BrITL sites if RPA is excluded. Shastri et al. also extended their BrITL analysis to the human triple-negative breast cancer cell line MDA-MA-231 (Shastri et al., 2018). In human cells treated with ATR inhibitor and aphidicolin, BrITL sites overlap with the open chromatin mark H3K4me3 and common fragile sites (CFSs), and they often contain repetitive sequences near the center. However, in contrast to the RPLs and BrITL sites in mouse cells, many human BrITL sites are AT-rich. A number of inverted transposon elements and microsatellites were found at human BrITL sites, and they are predicted to form quasi- or perfect palindromes (Figure 1). Thus, while the difficult-toreplicate sites in human and mouse genomes do not share sequence similarities, they are commonly associated with repetitive sequences that form stable secondary structures. The DNA sequences with these features may be a significant contributor to replication stress in the genome. Similar to the study by Shastri et al., Tubbs et al. also sought to identify sites
Molecular Cell 72, October 18, 2018 ª 2018 Elsevier Inc. 201
Molecular Cell
Previews
Figure 1. Mapping of Difficult-to-Replicate Sites in the Genome with Three Genomics Approaches Left panel: RPA ChIP was used by Shastri et al. to map the sites of stalled/collapsed replication forks in mouse cells treated with ATR inhibitor and aphidicolin. The sites identified by RPA ChIP are enriched for purine-rich microsatellite repeats. Central panel: BrITL was used by Shastri et al. to map the sites of DSBs induced by ATR inhibitor and aphidicolin in mouse and human cells. In addition to purine-rich microsatellites, BrITL identified inverted transposon elements and AT-rich microsatellites that form hairpins. Right panel: END-seq was used by Tubbs et al. to map the sites of hydroxyureainduced DSBs in early S-phase mouse cells. The sites identified by END-seq were enriched for poly(dA:dT) tracts.
of collapsed replication fork in the genome, but they did it in a different way (Tubbs et al., 2018). They treated mouse splenic B cells synchronously entering S phase with hydroxyurea (HU), an inhibitor of dNTP synthesis, to induce replication stress. The sites of recurrent DSBs in these early S-phase cells were identified by END-seq, a DNA end-mapping method with nucleotide resolution (Canela et al., 2016). Using this approach, they detected over 40,000 peaks of HUinduced DSBs in more than 6,000 zones (Figure 1). When cells were treated with a high concentration of HU (10 mM), the DSBs were enriched near the sites of replication initiation but excluded from transcribed regions. When a low concentration of HU (0.5 mM) was used, replication forks progressed into transcribed regions, generating DSBs throughout the replicated genome. In-depth analysis of the DSB sites uncovered that DSBs were formed in poly(dA:dT) tracts (Figure 1). Poly(dA:dT) tracts were not only found at HU-induced DSBs, but also at CFSs, spontaneous DSBs, and ATR inhibitor-induced DSBs, suggesting that they are a common cause of DSBs. The END-seq analysis by Tubbs et al. also revealed important details of how replication forks collapse at poly(dA:dT) 202 Molecular Cell 72, October 18, 2018
tracts. In contrast to the models predicting one-ended DSBs at collapsed replication forks (Sakofsky and Malkova, 2017), END-seq detected two DNA ends at HU-induced and spontaneous DSBs. One of the DNA ends is highly enriched around the tenth dT from the 30 ends of poly(dT), suggesting that the leadingstrand DNA polymerase is sharply stalled at this position when it runs into poly(dT) tracts. The second DNA ends of the DSBs are 250–400 nt away from the first ends. This observation is consistent with the possibility that the uncoupling of replicative helicase and DNA polymerase at stalled forks leads to fork breakage ahead of polymerase, leaving three DNA ends at two positions 250–400 nt apart after the ssDNA between the helicase and the polymerase is removed during the processing of END-seq samples (Figure 1). There are a number of noticeable differences between the two studies discussed. The study by Shastri et al. highlights the role of repetitive sequences in forming secondary structures that impose replication stress. In contrast, the study by Tubbs et al. suggests that poly(dA:dT) tracts are widespread barriers of replication forks. It is important to note that these two studies used
different cell types, growth/synchronization conditions, replication stress inducers, treatment durations, and methods to map collapsed forks or DSBs. Some of these technical differences could account for the different findings. It is also important to note that there are interesting similarities between the findings of these two studies. For example, both studies suggest that DNA sequences that stall DNA polymerase promote DSB formation. Both studies found evidence that AT-rich sequence elements are associated with fork-derived DSBs, although different sequence preferences were suggested. The replication stalling sequences discovered by both studies may give rise to non-B-form DNA structures. Finally, both studies found evidence for RPA exclusion from ssDNA at some sites of fork-derived DSBs. Both the unique and common findings of these two studies provide a new basis to understand how replication stress arises from DNA sequences in the genome. In future studies, it will be important to unravel how various DNA sequence elements contribute to replication stress in various physiological, pathological, and therapeutic contexts. ACKNOWLEDGMENTS H.D.N. is partly supported by NIH T32 postdoctoral training grant (T32 DK007540). L.Z. is the James & Patricia Poitras Endowed Chair in Cancer Research, and a Jim & Ann Orr Massachusetts General Hospital Research Scholar. L.Z. is supported by grants from the NIH (GM076388, CA197779, and CA218856). REFERENCES Buisson, R., Boisvert, J.L., Benes, C.H., and Zou, L. (2015). Distinct but concerted roles of ATR, DNA-PK, and Chk1 in countering replication stress during S phase. Mol. Cell 59, 1011–1024. Canela, A., Sridharan, S., Sciascia, N., Tubbs, A., Meltzer, P., Sleckman, B.P., and Nussenzweig, A. (2016). DNA breaks and end resection measured genome-wide by end sequencing. Mol. Cell 63, 898–911. Casper, A.M., Nghiem, P., Arlt, M.F., and Glover, T.W. (2002). ATR regulates fragile site stability. Cell 111, 779–789. Glover, T.W., Wilson, T.E., and Arlt, M.F. (2017). Fragile sites in cancer: more than meets the eye. Nat. Rev. Cancer 17, 489–501. Neil, A.J., Kim, J.C., and Mirkin, S.M. (2017). Precarious maintenance of simple DNA repeats in
Molecular Cell
Previews eukaryotes. BioEssays 39, https://doi.org/10. 1002/bies.201700077. Sakofsky, C.J., and Malkova, A. (2017). Break induced replication in eukaryotes: mechanisms, functions, and consequences. Crit. Rev. Biochem. Mol. Biol. 52, 395–413. Saldivar, J.C., Cortez, D., and Cimprich, K.A. (2017). The essential kinase ATR: ensuring faithful
duplication of a challenging genome. Nat. Rev. Mol. Cell Biol. 18, 622–636. Shastri, N., Tsai, Y., Hile, S., Jordan, D., Powell, B., Chen, J., Maloney, D., Dose, M., Lo, Y., Anaslassiadis, T., et al. (2018). Genome-wide identification of structure-forming repeats as principal sites of fork collapse upon ATR inhibition. Mol. Cell 72, this issue, 222–238.
Tubbs, A., Sridharan, S., van Wietmarschen, N., Maman, Y., Callen, E., Stanlie, A., Wu, W., Wu, X., Day, A., Wong, N., et al. (2018). Dual roles of poly(dA:dT) tracts in replication initiation and fork collapse. Cell 174, 1127–1142.e19. Zou, L., and Elledge, S.J. (2003). Sensing DNA damage through ATRIP recognition of RPA-ssDNA complexes. Science 300, 1542–1548.
WIPIng the Brakes off Autophagy Induction Timothy Marsh1,2 and Jayanta Debnath1,* 1Department of Pathology and Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, CA 94143, USA 2Biomedical Sciences Graduate Program, University of California, San Francisco, San Francisco, CA 94143, USA *Correspondence: [email protected] https://doi.org/10.1016/j.molcel.2018.10.003
In this issue of Molecular Cell, Wan et al. (2018) uncover WIPI2 as a critical rheostat in the control of autophagic degradation by mTORC1 and demonstrate the physiological utility of this signaling axis in promoting the clearance of hepatic lipids. Macroautophagy, hereafter termed autophagy, is an intracellular catabolic process involving the sequestration of cytosolic proteins and organelles into double-membrane vesicles, or autophagosomes, which are subsequently delivered to the lysosome for degradation. Due to the critical role of autophagy under homeostatic conditions across cell types as well as in diseases such as cancer, neurodegeneration, and aging, there is significant interest in identifying the key factors controlling the rate and specificity of autophagic degradation (Kroemer, 2015). One of the first discovered upstream regulators of autophagy was the mammalian target of rapamycin complex 1 (mTORC1) (Yang and Klionsky, 2010). mTORC1 integrates intracellular and extracellular signals to govern cell growth and metabolism and activation of this signaling axis inhibits autophagic degradation and recycling. In the current issue of Molecular Cell, Wan et al. (2018) identify WIPI2, a WD40domain-containing protein that functions as the mammalian ortholog of yeast Atg18, as a novel mTORC1 substrate that acts as a rheostat controlling the amplitude of autophagic induction downstream of mTORC1.
By integrating upstream signals, including nutrient levels, amino acid levels, and growth factor stimulation, mTORC1 directs whether cells grow and build biomass versus recycle cellular material through autophagy. Under nutrient-rich conditions, mTORC1 coordinately suppresses autophagy at several steps, including preventing autophagic membrane nucleation by the core autophagy machinery (i.e., ULK1, ATG13, and ATG14L), as well as by inhibiting both autophagosomal maturation and autolysosomal tubulation through the regulation of UVRAG (Chang and Neufeld, 2009; Kim et al., 2015; Yuan et al., 2013). Furthermore, TFEB-mediated transcription of lysosomal and autophagosomal genes sustains autophagic responses under prolonged mTORC1 inhibition (Roczniak-Ferguson et al., 2012). Upon nutrient starvation or inhibition of mTORC1, one can predict that the amplitude or intensity of autophagic degradation must be tuned to the particular needs and environment of the cell. Moreover, these signals that modulate the degree of autophagic flux may consequently influence the spectrum of substrates degraded by the autophagy pathway. Wan et al. (2018) identify mTORC1 phosphorylation
of Serine 395 on WIPI2 as one such key signal. mTORC1 phosphorylation of WIPI2 governs the ability of the E3 ubiquitin ligase, HUWE1, to target WIPI2 for proteasomal degradation, thereby attenuating autophagy. Overall, by uncovering these previously unrecognized signaling mechanisms and post-translational modifications controlling WIPI2 protein stabilization, Wan et al. (2018) provide new molecular insight into how the levels of WIPI2 protein finetune autophagic flux in cells both in vitro and in vivo (Figure 1). By generating a site-specific phosphorylation-resistant WIPI2 mutant (WIPI2S395A), Wan et al. (2018) uncouple the broader effects of mTORC1 on the core autophagy machinery from mTORC1WIPI2-mediated control of autophagic intensity. Importantly, enforcing WIPI2S395A expression enhances autophagic flux (Figure 1). Remarkably, overexpressing wild-type WIPI2 is by itself sufficient to boost autophagic flux, indicating that increasing WIPI2 protein abundance can supersede mTORC1 and HUWE1 regulation of WIPI2. On the whole, these studies illuminate novel strategies to enhance autophagy and hence interrogate the cellular consequences of augmented
Molecular Cell 72, October 18, 2018 ª 2018 Elsevier Inc. 203
Previews Getting a Genomic View of DNA Replication Stress Lee Zou1,2,* and Hai Dang Nguyen1 1Massachusetts
General Hospital Cancer Center, Harvard Medical School, Charlestown, MA 02129, USA of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02115 *Correspondence: [email protected] https://doi.org/10.1016/j.molcel.2018.10.001 2Department
DNA replication forks collapse at numerous sites throughout the genome under replication stress. Studies by Shastri et al. (2018) and Tubbs et al. (2018) used different genomics approaches to map the sites of replication fork collapse, revealing the contribution of specific DNA sequences to replication stress.
DNA replication stress is a major cause of genomic instability in the cell (Saldivar et al., 2017). Mounting evidence has shown that DNA replication stress arises from a variety of sources, such as dysregulation of replication initiation, defects in replication forks, compromised DNA repair or DNA damage signaling, and conflicts between replication and transcription. A number of chromosomal loci and DNA sequence elements have been implicated in the generation of replication stress and genomic instability (Glover et al., 2017; Neil et al., 2017). However, a genomic view of the major sites that impose replication stress is still lacking. Two independent studies by the laboratories of Brown and Nussenzweig have used different approaches to map the difficult-to-replicate sites in the genome, revealing a widespread contribution of specific DNA sequence elements to replication stress (Shastri et al., 2018; Tubbs et al., 2018). DNA replication forks encountering impediments on DNA commonly expose more single-stranded DNA (ssDNA), which is rapidly coated by the ssDNA-binding protein RPA. The RPA-coated ssDNA at stressed replication forks recruits the ATR-ATRIP kinase complex and promotes its activation, enabling ATR to stabilize stressed replication forks (Zou and Elledge, 2003). When ATR is inhibited, replication origin firing is increased and replication forks become prone to stalling and collapse at sites difficult to replicate, which may lead to RPA accumulation at these sites (Buisson et al., 2015; Casper et al., 2002). Using RPA accumulation as a surrogate for stalled/ collapsed replication forks, Shastri et al.
performed RPA chromatin immunoprecipitation sequencing (ChIP-seq) in mouse embryonic fibroblasts (MEFs) treated with ATR inhibitor and aphidicolin, an inhibitor of DNA polymerases, and mapped 171 ATR-dependent replication perturbed locations (RPLs) in the genome (Shastri et al., 2018) (Figure 1). Careful analysis of the RPLs revealed several common sequence features. Many RPLs contain microsatellite repeats near the center. These short tandem repeats are enriched for purines and specific DNA sequences, such as (CAGAGG/ and (CACAG/CTGTG)n CCTCTG)n (Figure 1). When analyzed in vitro, synthetic oligos of these repeats displayed features of secondary DNA structures and non-B-form DNA. Furthermore, the (CCTCTG)n repeats impeded the DNA synthesis by polymerase d in vitro and increased replication fork stalling on plasmids in cells. Collectively, these results suggest that the purine-rich microsatellite repeats in RPLs are natural barriers of replication forks. In addition to mapping RPLs with RPA ChIP-seq, Shastri et al. used another method called BrITL (breaks identified by TdT labeling) to identify the sites of DNA double-stranded breaks (DSBs) in MEFs treated with ATR inhibitor and aphidicolin (Shastri et al., 2018) (Figure 1). A subset of BrITL sites coincides with RPLs, suggesting that some of the DSBs induced by ATR inhibitor and aphidicolin arise from replication forks stalled at RPLs. However, BrITL also detected many DSBs that did not recruit RPA. These RPL-independent BrITL sites commonly contain long inverted repeats and quasi-palindromes with high melting
temperatures. Although these BrITL sites are distinct from RPLs in sequences, they share similarities in the repetitive nature and potential to form secondary DNA structures. Whether the RPA-free BrITL sites stall replication forks as RPLs still needs to be investigated. If replication forks were stalled at the BrITL sites, it would be important to understand why RPA is excluded from some of these sites and whether RPA exclusion contributes to DNA breakage. Furthermore, it would be interesting to investigate how ATR protects these BrITL sites if RPA is excluded. Shastri et al. also extended their BrITL analysis to the human triple-negative breast cancer cell line MDA-MA-231 (Shastri et al., 2018). In human cells treated with ATR inhibitor and aphidicolin, BrITL sites overlap with the open chromatin mark H3K4me3 and common fragile sites (CFSs), and they often contain repetitive sequences near the center. However, in contrast to the RPLs and BrITL sites in mouse cells, many human BrITL sites are AT-rich. A number of inverted transposon elements and microsatellites were found at human BrITL sites, and they are predicted to form quasi- or perfect palindromes (Figure 1). Thus, while the difficult-toreplicate sites in human and mouse genomes do not share sequence similarities, they are commonly associated with repetitive sequences that form stable secondary structures. The DNA sequences with these features may be a significant contributor to replication stress in the genome. Similar to the study by Shastri et al., Tubbs et al. also sought to identify sites
Molecular Cell 72, October 18, 2018 ª 2018 Elsevier Inc. 201
Molecular Cell
Previews
Figure 1. Mapping of Difficult-to-Replicate Sites in the Genome with Three Genomics Approaches Left panel: RPA ChIP was used by Shastri et al. to map the sites of stalled/collapsed replication forks in mouse cells treated with ATR inhibitor and aphidicolin. The sites identified by RPA ChIP are enriched for purine-rich microsatellite repeats. Central panel: BrITL was used by Shastri et al. to map the sites of DSBs induced by ATR inhibitor and aphidicolin in mouse and human cells. In addition to purine-rich microsatellites, BrITL identified inverted transposon elements and AT-rich microsatellites that form hairpins. Right panel: END-seq was used by Tubbs et al. to map the sites of hydroxyureainduced DSBs in early S-phase mouse cells. The sites identified by END-seq were enriched for poly(dA:dT) tracts.
of collapsed replication fork in the genome, but they did it in a different way (Tubbs et al., 2018). They treated mouse splenic B cells synchronously entering S phase with hydroxyurea (HU), an inhibitor of dNTP synthesis, to induce replication stress. The sites of recurrent DSBs in these early S-phase cells were identified by END-seq, a DNA end-mapping method with nucleotide resolution (Canela et al., 2016). Using this approach, they detected over 40,000 peaks of HUinduced DSBs in more than 6,000 zones (Figure 1). When cells were treated with a high concentration of HU (10 mM), the DSBs were enriched near the sites of replication initiation but excluded from transcribed regions. When a low concentration of HU (0.5 mM) was used, replication forks progressed into transcribed regions, generating DSBs throughout the replicated genome. In-depth analysis of the DSB sites uncovered that DSBs were formed in poly(dA:dT) tracts (Figure 1). Poly(dA:dT) tracts were not only found at HU-induced DSBs, but also at CFSs, spontaneous DSBs, and ATR inhibitor-induced DSBs, suggesting that they are a common cause of DSBs. The END-seq analysis by Tubbs et al. also revealed important details of how replication forks collapse at poly(dA:dT) 202 Molecular Cell 72, October 18, 2018
tracts. In contrast to the models predicting one-ended DSBs at collapsed replication forks (Sakofsky and Malkova, 2017), END-seq detected two DNA ends at HU-induced and spontaneous DSBs. One of the DNA ends is highly enriched around the tenth dT from the 30 ends of poly(dT), suggesting that the leadingstrand DNA polymerase is sharply stalled at this position when it runs into poly(dT) tracts. The second DNA ends of the DSBs are 250–400 nt away from the first ends. This observation is consistent with the possibility that the uncoupling of replicative helicase and DNA polymerase at stalled forks leads to fork breakage ahead of polymerase, leaving three DNA ends at two positions 250–400 nt apart after the ssDNA between the helicase and the polymerase is removed during the processing of END-seq samples (Figure 1). There are a number of noticeable differences between the two studies discussed. The study by Shastri et al. highlights the role of repetitive sequences in forming secondary structures that impose replication stress. In contrast, the study by Tubbs et al. suggests that poly(dA:dT) tracts are widespread barriers of replication forks. It is important to note that these two studies used
different cell types, growth/synchronization conditions, replication stress inducers, treatment durations, and methods to map collapsed forks or DSBs. Some of these technical differences could account for the different findings. It is also important to note that there are interesting similarities between the findings of these two studies. For example, both studies suggest that DNA sequences that stall DNA polymerase promote DSB formation. Both studies found evidence that AT-rich sequence elements are associated with fork-derived DSBs, although different sequence preferences were suggested. The replication stalling sequences discovered by both studies may give rise to non-B-form DNA structures. Finally, both studies found evidence for RPA exclusion from ssDNA at some sites of fork-derived DSBs. Both the unique and common findings of these two studies provide a new basis to understand how replication stress arises from DNA sequences in the genome. In future studies, it will be important to unravel how various DNA sequence elements contribute to replication stress in various physiological, pathological, and therapeutic contexts. ACKNOWLEDGMENTS H.D.N. is partly supported by NIH T32 postdoctoral training grant (T32 DK007540). L.Z. is the James & Patricia Poitras Endowed Chair in Cancer Research, and a Jim & Ann Orr Massachusetts General Hospital Research Scholar. L.Z. is supported by grants from the NIH (GM076388, CA197779, and CA218856). REFERENCES Buisson, R., Boisvert, J.L., Benes, C.H., and Zou, L. (2015). Distinct but concerted roles of ATR, DNA-PK, and Chk1 in countering replication stress during S phase. Mol. Cell 59, 1011–1024. Canela, A., Sridharan, S., Sciascia, N., Tubbs, A., Meltzer, P., Sleckman, B.P., and Nussenzweig, A. (2016). DNA breaks and end resection measured genome-wide by end sequencing. Mol. Cell 63, 898–911. Casper, A.M., Nghiem, P., Arlt, M.F., and Glover, T.W. (2002). ATR regulates fragile site stability. Cell 111, 779–789. Glover, T.W., Wilson, T.E., and Arlt, M.F. (2017). Fragile sites in cancer: more than meets the eye. Nat. Rev. Cancer 17, 489–501. Neil, A.J., Kim, J.C., and Mirkin, S.M. (2017). Precarious maintenance of simple DNA repeats in
Molecular Cell
Previews eukaryotes. BioEssays 39, https://doi.org/10. 1002/bies.201700077. Sakofsky, C.J., and Malkova, A. (2017). Break induced replication in eukaryotes: mechanisms, functions, and consequences. Crit. Rev. Biochem. Mol. Biol. 52, 395–413. Saldivar, J.C., Cortez, D., and Cimprich, K.A. (2017). The essential kinase ATR: ensuring faithful
duplication of a challenging genome. Nat. Rev. Mol. Cell Biol. 18, 622–636. Shastri, N., Tsai, Y., Hile, S., Jordan, D., Powell, B., Chen, J., Maloney, D., Dose, M., Lo, Y., Anaslassiadis, T., et al. (2018). Genome-wide identification of structure-forming repeats as principal sites of fork collapse upon ATR inhibition. Mol. Cell 72, this issue, 222–238.
Tubbs, A., Sridharan, S., van Wietmarschen, N., Maman, Y., Callen, E., Stanlie, A., Wu, W., Wu, X., Day, A., Wong, N., et al. (2018). Dual roles of poly(dA:dT) tracts in replication initiation and fork collapse. Cell 174, 1127–1142.e19. Zou, L., and Elledge, S.J. (2003). Sensing DNA damage through ATRIP recognition of RPA-ssDNA complexes. Science 300, 1542–1548.
WIPIng the Brakes off Autophagy Induction Timothy Marsh1,2 and Jayanta Debnath1,* 1Department of Pathology and Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, CA 94143, USA 2Biomedical Sciences Graduate Program, University of California, San Francisco, San Francisco, CA 94143, USA *Correspondence: [email protected] https://doi.org/10.1016/j.molcel.2018.10.003
In this issue of Molecular Cell, Wan et al. (2018) uncover WIPI2 as a critical rheostat in the control of autophagic degradation by mTORC1 and demonstrate the physiological utility of this signaling axis in promoting the clearance of hepatic lipids. Macroautophagy, hereafter termed autophagy, is an intracellular catabolic process involving the sequestration of cytosolic proteins and organelles into double-membrane vesicles, or autophagosomes, which are subsequently delivered to the lysosome for degradation. Due to the critical role of autophagy under homeostatic conditions across cell types as well as in diseases such as cancer, neurodegeneration, and aging, there is significant interest in identifying the key factors controlling the rate and specificity of autophagic degradation (Kroemer, 2015). One of the first discovered upstream regulators of autophagy was the mammalian target of rapamycin complex 1 (mTORC1) (Yang and Klionsky, 2010). mTORC1 integrates intracellular and extracellular signals to govern cell growth and metabolism and activation of this signaling axis inhibits autophagic degradation and recycling. In the current issue of Molecular Cell, Wan et al. (2018) identify WIPI2, a WD40domain-containing protein that functions as the mammalian ortholog of yeast Atg18, as a novel mTORC1 substrate that acts as a rheostat controlling the amplitude of autophagic induction downstream of mTORC1.
By integrating upstream signals, including nutrient levels, amino acid levels, and growth factor stimulation, mTORC1 directs whether cells grow and build biomass versus recycle cellular material through autophagy. Under nutrient-rich conditions, mTORC1 coordinately suppresses autophagy at several steps, including preventing autophagic membrane nucleation by the core autophagy machinery (i.e., ULK1, ATG13, and ATG14L), as well as by inhibiting both autophagosomal maturation and autolysosomal tubulation through the regulation of UVRAG (Chang and Neufeld, 2009; Kim et al., 2015; Yuan et al., 2013). Furthermore, TFEB-mediated transcription of lysosomal and autophagosomal genes sustains autophagic responses under prolonged mTORC1 inhibition (Roczniak-Ferguson et al., 2012). Upon nutrient starvation or inhibition of mTORC1, one can predict that the amplitude or intensity of autophagic degradation must be tuned to the particular needs and environment of the cell. Moreover, these signals that modulate the degree of autophagic flux may consequently influence the spectrum of substrates degraded by the autophagy pathway. Wan et al. (2018) identify mTORC1 phosphorylation
of Serine 395 on WIPI2 as one such key signal. mTORC1 phosphorylation of WIPI2 governs the ability of the E3 ubiquitin ligase, HUWE1, to target WIPI2 for proteasomal degradation, thereby attenuating autophagy. Overall, by uncovering these previously unrecognized signaling mechanisms and post-translational modifications controlling WIPI2 protein stabilization, Wan et al. (2018) provide new molecular insight into how the levels of WIPI2 protein finetune autophagic flux in cells both in vitro and in vivo (Figure 1). By generating a site-specific phosphorylation-resistant WIPI2 mutant (WIPI2S395A), Wan et al. (2018) uncouple the broader effects of mTORC1 on the core autophagy machinery from mTORC1WIPI2-mediated control of autophagic intensity. Importantly, enforcing WIPI2S395A expression enhances autophagic flux (Figure 1). Remarkably, overexpressing wild-type WIPI2 is by itself sufficient to boost autophagic flux, indicating that increasing WIPI2 protein abundance can supersede mTORC1 and HUWE1 regulation of WIPI2. On the whole, these studies illuminate novel strategies to enhance autophagy and hence interrogate the cellular consequences of augmented
Molecular Cell 72, October 18, 2018 ª 2018 Elsevier Inc. 203