news and views
↑ Ddit4 ↑ CSFR1
npg
© 2015 Nature America, Inc. All rights reserved.
Kim Caesar/Nature Publishing Group
SCFA
?
SPF mice Recolonized mice SCFA-fed GF mice
GF mice Antibiotic-treated SPF mice
Figure 1 Communication from gut to brain regulates microglia. The typical morphology, territorial boundaries and molecular profile of microglia observed in mice living in standard, clean housing conditions (SPF; left) are changed in mice living in a GF environment (right). Microglia of GF mice display extended processes that encroach on each other’s territories and a gene expression profile more akin to that of immature cells (for example, upregulation of CSFR1 and Ddit4). Partial ablation of gut microbiota with antibiotics produces a microglial phenotype similar to the one observed in GF mice, while recolonization of GF mice or feeding with SCFA normalizes the microglial phenotype. GF, germ free; SPF, specific pathogen free; SCFA, short-chain fatty acids.
of crossing the blood-brain barrier that regulates microglia. Another intriguing hypothesis, however, arises from the recent finding that germ-free mice have increased blood-brain barrier permeability beginning in utero and continuing into adulthood12. Perhaps this could mean that splenic myeloid cells, which do not normally contribute to the resident microglial population7,13, are able to enter the brain parenchyma in the absence of microbiota and SCFA, and there display characteristics of immature and impaired myeloid cells. The work by Erny et al.5 thus opens several new avenues for future research. These findings
clearly have important implications for human conditions in which the constitution of gut bacteria may be altered, such as ulcerative colitis, Crohn’s disease and irritable bowel syndrome14, or in which the bacteria are depleted, as happens during oral antibiotic use15. On this note, the researchers found that depleting the intestinal microbes of SPF mice during adulthood with antibiotics was sufficient to alter the morphology of microglia, such that they resembled the cells found in the brains of the GF mice that had never been exposed to complex microbiota. Though this highlights the sensitivity, and possible dysregulation, of
the gut-brain communication system, on a positive note, this work also demonstrates that some treatment may be possible in the form of bacterial reconstitution or SCFA, at least to alleviate the effects on microglia. In regard to basic biology, this paper provides a new perspective on the regulation of microglial development and function at a systemic level. Still more generally, this is also an exciting example of “developmental programming”2, showing how early environmental conditions, be they external or, in the special case of the gut microbiome, internal, influence the development of an organ. With studies like this continually demonstrating the link between microbiota and the brain, and with the observation that microglia can sculpt synaptic circuits, perhaps there is biological credibility to the concept of gut instincts. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. 1. Ley, R.E. et al. Science 320, 1647–1651 (2008). 2. Diaz Heijtz, R. et al. Proc. Natl. Acad. Sci. USA 108, 3047–3052 (2011). 3. Clarke, G. et al. Mol. Psychiatry 18, 666–673 (2013). 4. Mayer, E.A. Nat. Rev. Neurosci. 12, 453–466 (2011). 5. Erny, D. et al. Nat. Neurosci. 18, 965–977 (2015). 6. Kierdorf, K. et al. Nat. Neurosci. 16, 273–280 (2013). 7. Ginhoux, F. et al. Science 330, 841–845 (2010). 8. Kettenmann, H., Hanisch, U.-K., Noda, M. & Verkhratsky, A. Physiol. Rev. 91, 461–553 (2011). 9. Nikodemova, M. et al. J. Neuroimmunol. 278, 280–288 (2015). 10. Perry, V.H., Hume, D.A. & Gordon, S. Neuroscience 15, 313–326 (1985). 11. Schafer, D.P. et al. Neuron 74, 691–705 (2012). 12. Braniste, V. et al. Sci. Transl. Med. 6, 263ra158 (2014). 13. Mildner, A. et al. Nat. Neurosci. 10, 1544–1553 (2007). 14. Grenham, S., Clarke, G., Cryan, J.F. & Dinan, T.G. Frontiers Physiol. 2, 94 (2011). 15. Ng, K.M. et al. Nature 502, 96–99 (2013).
MIR137: big impacts from small changes Jinju Han, Anindita Sarkar & Fred H Gage Schizophrenia-linked single nucleotide polymorphisms in MIR137 alter expression of miR-137 in neurons. Abnormal expression of miR-137 affects vesicle release at presynaptic terminals and in turn alters hippocampal functioning. Schizophrenia (SZ) is a chronic psychiatric disorder presenting diverse symptoms, including hallucinations, delusions and memory impairment. SZ affects around 1% Jinju Han, Anindita Sarkar and Fred H. Gage are in the Laboratory of Genetics, the Salk Institute for Biological Studies, La Jolla, California, USA. e-mail: [email protected]
of the population globally. However, among individuals who have a first-degree relative with SZ, risk for this mental illness increases tenfold. Moreover, the concordance rate for SZ in monozygotic twins has been reported to be ~50%, indicating that the disease is strongly affected by heritable components1. This finding has driven researchers to identify genes that contribute to its etiology.
nature neuroscience volume 18 | number 7 | july 2015
Genome-wide association studies have reported genomic regions with copy number variations (CNVs) and single nucleotide polymorphisms (SNPs) that are potentially linked to SZ2,3. However, it is not yet fully understood how the genomic variations are related to symptoms of SZ. In this issue of Nature Neuroscience, Siegert et al.4 demonstrate the effects of SNPs in a SZ-associated 931
news and views
TFs? SNPs
? MIR137
?
p(A)
?
?
miR-137
CPLX1, NSF, SYN3, SYT1, . . .
MIR137
p(A)
miR-137
npg
© 2015 Nature America, Inc. All rights reserved.
CPLX1, NSF, SYN3, SYT1, . . .
Figure 1 Enhanced expression of miR-137 restrains release of presynaptic vesicles. SNPs in MIR137 that are associated with the onset of SZ may recruit different transcription factors (TFs) and alter transcription of the gene. Induced neurons derived from subjects with SZ express more miR-137 than induced neurons from healthy controls. The increased expression of miR-137 reduces the generation of proteins working on vesicle trafficking, such as CPLX1, NSF, SYN3 and SYT1. This reduction in vesicle-trafficking proteins results in delayed release of presynaptic vesicles and a consequent decrease in postsynaptic activity, including LTP generation. The physiological relevance of this sequential process is demonstrated in the mossy fibers of the mouse hippocampus. The expression level of miR-137 in the dentate gyrus is correlated with hippocampus-dependent learning and memory.
gene, MIR137, on presynaptic functions of neurons (Fig. 1). The MIR137 gene, which encodes microRNA-137 (miR-137), has attracted attention because SNPs in the flanking regions of the gene are highly associated with increased SZ risk and many genes in CNV regions associating with SZ are predicted to be downstream of miR-137 (refs. 3,5). The effects of SNPs on miR-137 expression have been investigated by measuring the expression of endogenous miR-137 in the postmortem brain6 and the activity of reporter genes possessing the SNPs in neuron-like cell lines7. However, previous studies have not described the effects of the SNPs on endogenous miR-137 expression in neurons specifically. The expression and cellular functions of microRNAs are regulated in a cell type– dependent manner8. Siegert et al.4 derived induced neurons from fibroblasts of subjects with SZ carrying minor alleles of SNPs in MIR137 to study the neuron-specific roles of miR-137. The authors found that miR137 expression was increased specifically in neurons, but not fibroblasts, derived from subjects with SZ, relative to expression in the corresponding cell types derived from healthy controls. 932
microRNAs regulate expression of proteincoding genes at the post-transcriptional level, and their target genes have been predicted by computational analyses9. To explore the functional relevance of the elevated miR-137 in neurons of SZ patients, Siegert et al.4 performed in silico analysis on predicted genes from multiple databases to find target genes that were regulated by miR-137 in neurons. A gene ontology analysis of the screened genes revealed that a subset of genes is involved in synaptic transmission. The authors narrowed down the list to four genes, CPLX1, NSF, SYN3 and SYT1, that have well-established roles in presynaptic vesicle trafficking at the synapse and are directly regulated by miR-137 in neurons. Induced neurons derived from subjects with SZ indeed expressed these four genes at lower levels than those from controls, indicating that elevated miR-137 represses them. Reflecting these results, activitydependent release of FM4-64, a marker for vesicle trafficking, was delayed in induced neurons from subjects with SZ compared with controls4. Siegert et al.4 extended their study to mouse hippocampus to elucidate the functional relevance of altered expression of miR-137 in vivo. There is mounting evidence for
dysfunction in the hippocampal circuitry of patients with SZ10. For instance, synaptic density at the mossy fiber terminals in brains from subjects with SZ is reduced11. To recapitulate conditions in the induced human neurons from subjects with SZ, which expressed more miR-137, the authors overexpressed miR-137 in the dentate gyrus using a lentiviral vector. They confirmed that this manipulation suppressed all four vesicle-trafficking genes in the mossy fibers, as observed in the induced human neurons. They then visualized distributions of presynaptic vesicles at the mossy fibers through electron microscopy. Overexpression of miR-137 decreased the number of releaseready vesicles from the active zones of mossy fiber presynaptic terminals. Overexpression of miR-137 also resulted in lower field excitatory postsynaptic potential (fEPSP) amplitude and less pronounced long-term potentiation (LTP) at mossy fibers compared with those in controls. Siegert et al.4 further investigated the behavioral effects of overexpression of miR-137 to determine the implications of these cellular effects on hippocampal function. Animals overexpressing miR-137 in the dentate gyrus displayed less freezing behavior in a contextual fear conditioning test and performed poorly on a Morris water maze test; however, they did not show any differences in cued conditioning, locomotor activity or anxiety tests. These data suggest that SZ patients harboring minor alleles of SNPs in MIR137 may have impairment of hippocampus-dependent functions. Siegert et al.4 also performed loss-offunction studies in both induced neurons from subjects with SZ and the mouse hippocampus. Three of the four vesicle-trafficking target genes (all except Syn3) were upregulated when miR-137 was sequestered by a microRNA sponge in both model systems. Consistent with the upregulation of these three synaptic genes, when miR-137 was suppressed by the sponge, efficiency of vesicle trafficking at the presynaptic terminals was improved in both induced neurons and the mouse hippocampus. Increased fEPSP and LTP in mouse hippocampus expressing the microRNA sponge also argues for enhanced synaptic activity at mossy fibers. Interestingly, reducing miR-137 alone was sufficient to improve hippocampusdependent learning in the contextual fear conditioning test. Evidence for mechanistic regulation came with a rescue experiment in which Syt1 and miR-137 were coexpressed in the hippocampus. Overexpressing Syt1 alone with miR-137 was insufficient to completely restore the effects of miR-137 overexpression on the presynaptic deficit, as miR-137 regulates several genes. However, Syt1 did
volume 18 | number 7 | july 2015 nature neuroscience
npg
© 2015 Nature America, Inc. All rights reserved.
news and views partially counteract miR-137 and compensate to a considerable extent for presynaptic dysfunctions caused by miR-137 overexpression. All these results affirm a functional effect of miR-137 on presynaptic terminals regarding vesicle trafficking. This study by Siegert et al.4 is the first indepth characterization of the functions and molecular mechanisms of SZ-associated common genomic variants in MIR137. The real strength of the study lies in how the authors tie together genetic, molecular and behavioral data to shed light on the role of miR-137 in SZ. It is also interesting that the authors did similar experiments in both induced neurons and the mouse hippocampus and that each model system contributed unique mechanistic insights. For example, presynaptic vesicular trafficking done in the induced neurons was corroborated by the relevant behavior experiments in mice. The experiments in the mouse hippocampus were particularly
interesting as they offered a very nice system in which to study the full mechanistic and functional effect of SZ-associated SNPs. However, the question of whether the abnormal vesicle trafficking mediated by the minor SNPs represents a general mechanism leading to SZ pathology remains to be addressed. It would be important to know whether miR137 causes a similar dysfunction in other brain areas commonly affected by SZ, such as the prefrontal cortex. Nevertheless, the conserved phenotypes in the mouse hippocampus and induced neurons are encouraging, holding the promise of finding a marker or assay for SZ with predictive diagnostic value using patient-derived induced neurons. We must, however, take the phenotypic results in this study with caution, as the induced neurons were derived from only two individuals carrying minor SNPs. Future studies addressing this topic would benefit from a bigger cohort encompassing diverse genetic
backgrounds or monozygotic twins with a homogeneous background who are discordant for SZ. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. 1. Picchioni, M.M. & Murray, R.M. Br. Med. J. 335, 91–95 (2007). 2. Szatkiewicz, J.P. et al. Mol. Psychiatry 19, 762–773 (2014). 3. Schizophrenia Working Group of the Psychiatric Genomics Consortium. Nature 511, 421–427 (2014). 4. Siegert, S. et al. Nat. Neurosci. 18, 1008–1016 (2015). 5. Ripke, S. et al. Nat. Genet. 43, 969–976 (2011). 6. Guella, I. et al. J. Psychiatr. Res. 47, 1215–1221 (2013). 7. Strazisar, M. et al. Mol. Psychiatry 20, 472–481 (2015). 8. Erhard, F. et al. Genome Res. 24, 906–919 (2014). 9. Bartel, D.P. Cell 136, 215–233 (2009). 10. Harrison, P.J. Psychopharmacology (Berl.) 174, 151–162 (2004). 11. Kolomeets, N.S., Orlovskaya, D.D. & Uranova, N.A. Synapse 61, 615–621 (2007).
How amyloid, sleep and memory connect Brendan P Lucey & David M Holtzman In a bidirectional relationship, the sleep/wake cycle regulates amyloid-b (Ab) levels and Ab accumulation then disrupts sleep. A quantitative three-way model now suggests that Ab impairs memory via its effect on sleep. Alzheimer’s disease (AD) pathology, characterized by Aβ deposition in the brain as insoluble extracellular plaques and intracellular tau aggregation in paired helical filaments, begins to develop ~10–15 years before the onset of memory impairment1. By the time cognitive symptoms and signs are present, substantial synaptic and neuronal injury has already occurred. In recent years, one focus of AD research has been to define the preclinical phase of AD, when Aβ with or without neocortical tau deposition has begun to occur, but before clear cognitive decline. The ultimate goal is therapeutic intervention at this stage to prevent progression to symptomatic AD2. AD pathology has long been associated with memory impairment in older adults, but recent evidence has also shown Aβ deposition to be associated with disruption in sleep quality even in the absence of cognitive impairment3. Furthermore, recent evidence supports a role Brendan P. Lucey and David M. Holtzman are in the Department of Neurology, Hope Center for Neurological Disorders, Knight Alzheimer’s Disease Research Center, Washington University, St. Louis, Missouri, USA. e-mail: [email protected]
for sleep in the development of AD, at least in part by influencing Aβ. Aβ fluctuates diurnally: soluble Aβ levels are higher during wakefulness and lower during sleep4,5. Sleep deprivation accelerates Aβ deposition in APP transgenic mice4, whereas orexin deficiency, which increases sleep, decreases it6. In addition, amyloid deposition disrupts sleep in APP transgenic mice7. The relationship between sleep and Aβ deposition has thus been proposed to be bidirectional: sleep disruption leads to protein deposits and protein deposits result in sleep disturbance8. The relationship among Aβ deposition, other aspects of AD pathology, sleep and memory impairment are not well defined in humans. In this issue of Nature Neuroscience, Mander et al.9 report a link between brain Aβ deposition, sleep and memory dysfunction. They hypothesize that Aβ accumulation in the medial prefrontal cortex (mPFC) is associated with diminished slow-wave activity (SWA) during non-REM (NREM) sleep that further correlates with the extent of impaired overnight hippocampus-dependent memory consolidation in older adults. Previous work from this group has shown that mPFC atrophy is associated with reduced NREM SWA
nature neuroscience volume 18 | number 7 | july 2015
and that this association correlates with overnight memory retention10. To further investigate whether amyloid deposition rather than brain atrophy has a similar effect, the authors recruited 26 cognitively normal older adults who underwent positron emission tomography imaging with Pittsburgh compound B to determine the amount of fibrillar Aβ deposited in the brain. To assess memory function, all participants trained on a set of word pairs in the evening. Then sleep was monitored overnight with polysomnography to assess different sleep stages, such as NREM sleep, and to obtain electroencephalography (EEG) for power analysis. Participants took the word-pair test again in the morning during a functional MRI scanning session. The authors found that higher amyloid burden in the mPFC correlated with decreased NREM SWA in this brain region, but not with higher frequencies of EEG activity or with decreased NREM SWA in other regions. This decrease in mPFC NREM SWA further correlated with worse overnight memory retention, even after controlling for age and sex. The authors then sought to determine the interaction of these factors using path analysis. Three models were constructed to determine the nature of 933
↑ Ddit4 ↑ CSFR1
npg
© 2015 Nature America, Inc. All rights reserved.
Kim Caesar/Nature Publishing Group
SCFA
?
SPF mice Recolonized mice SCFA-fed GF mice
GF mice Antibiotic-treated SPF mice
Figure 1 Communication from gut to brain regulates microglia. The typical morphology, territorial boundaries and molecular profile of microglia observed in mice living in standard, clean housing conditions (SPF; left) are changed in mice living in a GF environment (right). Microglia of GF mice display extended processes that encroach on each other’s territories and a gene expression profile more akin to that of immature cells (for example, upregulation of CSFR1 and Ddit4). Partial ablation of gut microbiota with antibiotics produces a microglial phenotype similar to the one observed in GF mice, while recolonization of GF mice or feeding with SCFA normalizes the microglial phenotype. GF, germ free; SPF, specific pathogen free; SCFA, short-chain fatty acids.
of crossing the blood-brain barrier that regulates microglia. Another intriguing hypothesis, however, arises from the recent finding that germ-free mice have increased blood-brain barrier permeability beginning in utero and continuing into adulthood12. Perhaps this could mean that splenic myeloid cells, which do not normally contribute to the resident microglial population7,13, are able to enter the brain parenchyma in the absence of microbiota and SCFA, and there display characteristics of immature and impaired myeloid cells. The work by Erny et al.5 thus opens several new avenues for future research. These findings
clearly have important implications for human conditions in which the constitution of gut bacteria may be altered, such as ulcerative colitis, Crohn’s disease and irritable bowel syndrome14, or in which the bacteria are depleted, as happens during oral antibiotic use15. On this note, the researchers found that depleting the intestinal microbes of SPF mice during adulthood with antibiotics was sufficient to alter the morphology of microglia, such that they resembled the cells found in the brains of the GF mice that had never been exposed to complex microbiota. Though this highlights the sensitivity, and possible dysregulation, of
the gut-brain communication system, on a positive note, this work also demonstrates that some treatment may be possible in the form of bacterial reconstitution or SCFA, at least to alleviate the effects on microglia. In regard to basic biology, this paper provides a new perspective on the regulation of microglial development and function at a systemic level. Still more generally, this is also an exciting example of “developmental programming”2, showing how early environmental conditions, be they external or, in the special case of the gut microbiome, internal, influence the development of an organ. With studies like this continually demonstrating the link between microbiota and the brain, and with the observation that microglia can sculpt synaptic circuits, perhaps there is biological credibility to the concept of gut instincts. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. 1. Ley, R.E. et al. Science 320, 1647–1651 (2008). 2. Diaz Heijtz, R. et al. Proc. Natl. Acad. Sci. USA 108, 3047–3052 (2011). 3. Clarke, G. et al. Mol. Psychiatry 18, 666–673 (2013). 4. Mayer, E.A. Nat. Rev. Neurosci. 12, 453–466 (2011). 5. Erny, D. et al. Nat. Neurosci. 18, 965–977 (2015). 6. Kierdorf, K. et al. Nat. Neurosci. 16, 273–280 (2013). 7. Ginhoux, F. et al. Science 330, 841–845 (2010). 8. Kettenmann, H., Hanisch, U.-K., Noda, M. & Verkhratsky, A. Physiol. Rev. 91, 461–553 (2011). 9. Nikodemova, M. et al. J. Neuroimmunol. 278, 280–288 (2015). 10. Perry, V.H., Hume, D.A. & Gordon, S. Neuroscience 15, 313–326 (1985). 11. Schafer, D.P. et al. Neuron 74, 691–705 (2012). 12. Braniste, V. et al. Sci. Transl. Med. 6, 263ra158 (2014). 13. Mildner, A. et al. Nat. Neurosci. 10, 1544–1553 (2007). 14. Grenham, S., Clarke, G., Cryan, J.F. & Dinan, T.G. Frontiers Physiol. 2, 94 (2011). 15. Ng, K.M. et al. Nature 502, 96–99 (2013).
MIR137: big impacts from small changes Jinju Han, Anindita Sarkar & Fred H Gage Schizophrenia-linked single nucleotide polymorphisms in MIR137 alter expression of miR-137 in neurons. Abnormal expression of miR-137 affects vesicle release at presynaptic terminals and in turn alters hippocampal functioning. Schizophrenia (SZ) is a chronic psychiatric disorder presenting diverse symptoms, including hallucinations, delusions and memory impairment. SZ affects around 1% Jinju Han, Anindita Sarkar and Fred H. Gage are in the Laboratory of Genetics, the Salk Institute for Biological Studies, La Jolla, California, USA. e-mail: [email protected]
of the population globally. However, among individuals who have a first-degree relative with SZ, risk for this mental illness increases tenfold. Moreover, the concordance rate for SZ in monozygotic twins has been reported to be ~50%, indicating that the disease is strongly affected by heritable components1. This finding has driven researchers to identify genes that contribute to its etiology.
nature neuroscience volume 18 | number 7 | july 2015
Genome-wide association studies have reported genomic regions with copy number variations (CNVs) and single nucleotide polymorphisms (SNPs) that are potentially linked to SZ2,3. However, it is not yet fully understood how the genomic variations are related to symptoms of SZ. In this issue of Nature Neuroscience, Siegert et al.4 demonstrate the effects of SNPs in a SZ-associated 931
news and views
TFs? SNPs
? MIR137
?
p(A)
?
?
miR-137
CPLX1, NSF, SYN3, SYT1, . . .
MIR137
p(A)
miR-137
npg
© 2015 Nature America, Inc. All rights reserved.
CPLX1, NSF, SYN3, SYT1, . . .
Figure 1 Enhanced expression of miR-137 restrains release of presynaptic vesicles. SNPs in MIR137 that are associated with the onset of SZ may recruit different transcription factors (TFs) and alter transcription of the gene. Induced neurons derived from subjects with SZ express more miR-137 than induced neurons from healthy controls. The increased expression of miR-137 reduces the generation of proteins working on vesicle trafficking, such as CPLX1, NSF, SYN3 and SYT1. This reduction in vesicle-trafficking proteins results in delayed release of presynaptic vesicles and a consequent decrease in postsynaptic activity, including LTP generation. The physiological relevance of this sequential process is demonstrated in the mossy fibers of the mouse hippocampus. The expression level of miR-137 in the dentate gyrus is correlated with hippocampus-dependent learning and memory.
gene, MIR137, on presynaptic functions of neurons (Fig. 1). The MIR137 gene, which encodes microRNA-137 (miR-137), has attracted attention because SNPs in the flanking regions of the gene are highly associated with increased SZ risk and many genes in CNV regions associating with SZ are predicted to be downstream of miR-137 (refs. 3,5). The effects of SNPs on miR-137 expression have been investigated by measuring the expression of endogenous miR-137 in the postmortem brain6 and the activity of reporter genes possessing the SNPs in neuron-like cell lines7. However, previous studies have not described the effects of the SNPs on endogenous miR-137 expression in neurons specifically. The expression and cellular functions of microRNAs are regulated in a cell type– dependent manner8. Siegert et al.4 derived induced neurons from fibroblasts of subjects with SZ carrying minor alleles of SNPs in MIR137 to study the neuron-specific roles of miR-137. The authors found that miR137 expression was increased specifically in neurons, but not fibroblasts, derived from subjects with SZ, relative to expression in the corresponding cell types derived from healthy controls. 932
microRNAs regulate expression of proteincoding genes at the post-transcriptional level, and their target genes have been predicted by computational analyses9. To explore the functional relevance of the elevated miR-137 in neurons of SZ patients, Siegert et al.4 performed in silico analysis on predicted genes from multiple databases to find target genes that were regulated by miR-137 in neurons. A gene ontology analysis of the screened genes revealed that a subset of genes is involved in synaptic transmission. The authors narrowed down the list to four genes, CPLX1, NSF, SYN3 and SYT1, that have well-established roles in presynaptic vesicle trafficking at the synapse and are directly regulated by miR-137 in neurons. Induced neurons derived from subjects with SZ indeed expressed these four genes at lower levels than those from controls, indicating that elevated miR-137 represses them. Reflecting these results, activitydependent release of FM4-64, a marker for vesicle trafficking, was delayed in induced neurons from subjects with SZ compared with controls4. Siegert et al.4 extended their study to mouse hippocampus to elucidate the functional relevance of altered expression of miR-137 in vivo. There is mounting evidence for
dysfunction in the hippocampal circuitry of patients with SZ10. For instance, synaptic density at the mossy fiber terminals in brains from subjects with SZ is reduced11. To recapitulate conditions in the induced human neurons from subjects with SZ, which expressed more miR-137, the authors overexpressed miR-137 in the dentate gyrus using a lentiviral vector. They confirmed that this manipulation suppressed all four vesicle-trafficking genes in the mossy fibers, as observed in the induced human neurons. They then visualized distributions of presynaptic vesicles at the mossy fibers through electron microscopy. Overexpression of miR-137 decreased the number of releaseready vesicles from the active zones of mossy fiber presynaptic terminals. Overexpression of miR-137 also resulted in lower field excitatory postsynaptic potential (fEPSP) amplitude and less pronounced long-term potentiation (LTP) at mossy fibers compared with those in controls. Siegert et al.4 further investigated the behavioral effects of overexpression of miR-137 to determine the implications of these cellular effects on hippocampal function. Animals overexpressing miR-137 in the dentate gyrus displayed less freezing behavior in a contextual fear conditioning test and performed poorly on a Morris water maze test; however, they did not show any differences in cued conditioning, locomotor activity or anxiety tests. These data suggest that SZ patients harboring minor alleles of SNPs in MIR137 may have impairment of hippocampus-dependent functions. Siegert et al.4 also performed loss-offunction studies in both induced neurons from subjects with SZ and the mouse hippocampus. Three of the four vesicle-trafficking target genes (all except Syn3) were upregulated when miR-137 was sequestered by a microRNA sponge in both model systems. Consistent with the upregulation of these three synaptic genes, when miR-137 was suppressed by the sponge, efficiency of vesicle trafficking at the presynaptic terminals was improved in both induced neurons and the mouse hippocampus. Increased fEPSP and LTP in mouse hippocampus expressing the microRNA sponge also argues for enhanced synaptic activity at mossy fibers. Interestingly, reducing miR-137 alone was sufficient to improve hippocampusdependent learning in the contextual fear conditioning test. Evidence for mechanistic regulation came with a rescue experiment in which Syt1 and miR-137 were coexpressed in the hippocampus. Overexpressing Syt1 alone with miR-137 was insufficient to completely restore the effects of miR-137 overexpression on the presynaptic deficit, as miR-137 regulates several genes. However, Syt1 did
volume 18 | number 7 | july 2015 nature neuroscience
npg
© 2015 Nature America, Inc. All rights reserved.
news and views partially counteract miR-137 and compensate to a considerable extent for presynaptic dysfunctions caused by miR-137 overexpression. All these results affirm a functional effect of miR-137 on presynaptic terminals regarding vesicle trafficking. This study by Siegert et al.4 is the first indepth characterization of the functions and molecular mechanisms of SZ-associated common genomic variants in MIR137. The real strength of the study lies in how the authors tie together genetic, molecular and behavioral data to shed light on the role of miR-137 in SZ. It is also interesting that the authors did similar experiments in both induced neurons and the mouse hippocampus and that each model system contributed unique mechanistic insights. For example, presynaptic vesicular trafficking done in the induced neurons was corroborated by the relevant behavior experiments in mice. The experiments in the mouse hippocampus were particularly
interesting as they offered a very nice system in which to study the full mechanistic and functional effect of SZ-associated SNPs. However, the question of whether the abnormal vesicle trafficking mediated by the minor SNPs represents a general mechanism leading to SZ pathology remains to be addressed. It would be important to know whether miR137 causes a similar dysfunction in other brain areas commonly affected by SZ, such as the prefrontal cortex. Nevertheless, the conserved phenotypes in the mouse hippocampus and induced neurons are encouraging, holding the promise of finding a marker or assay for SZ with predictive diagnostic value using patient-derived induced neurons. We must, however, take the phenotypic results in this study with caution, as the induced neurons were derived from only two individuals carrying minor SNPs. Future studies addressing this topic would benefit from a bigger cohort encompassing diverse genetic
backgrounds or monozygotic twins with a homogeneous background who are discordant for SZ. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. 1. Picchioni, M.M. & Murray, R.M. Br. Med. J. 335, 91–95 (2007). 2. Szatkiewicz, J.P. et al. Mol. Psychiatry 19, 762–773 (2014). 3. Schizophrenia Working Group of the Psychiatric Genomics Consortium. Nature 511, 421–427 (2014). 4. Siegert, S. et al. Nat. Neurosci. 18, 1008–1016 (2015). 5. Ripke, S. et al. Nat. Genet. 43, 969–976 (2011). 6. Guella, I. et al. J. Psychiatr. Res. 47, 1215–1221 (2013). 7. Strazisar, M. et al. Mol. Psychiatry 20, 472–481 (2015). 8. Erhard, F. et al. Genome Res. 24, 906–919 (2014). 9. Bartel, D.P. Cell 136, 215–233 (2009). 10. Harrison, P.J. Psychopharmacology (Berl.) 174, 151–162 (2004). 11. Kolomeets, N.S., Orlovskaya, D.D. & Uranova, N.A. Synapse 61, 615–621 (2007).
How amyloid, sleep and memory connect Brendan P Lucey & David M Holtzman In a bidirectional relationship, the sleep/wake cycle regulates amyloid-b (Ab) levels and Ab accumulation then disrupts sleep. A quantitative three-way model now suggests that Ab impairs memory via its effect on sleep. Alzheimer’s disease (AD) pathology, characterized by Aβ deposition in the brain as insoluble extracellular plaques and intracellular tau aggregation in paired helical filaments, begins to develop ~10–15 years before the onset of memory impairment1. By the time cognitive symptoms and signs are present, substantial synaptic and neuronal injury has already occurred. In recent years, one focus of AD research has been to define the preclinical phase of AD, when Aβ with or without neocortical tau deposition has begun to occur, but before clear cognitive decline. The ultimate goal is therapeutic intervention at this stage to prevent progression to symptomatic AD2. AD pathology has long been associated with memory impairment in older adults, but recent evidence has also shown Aβ deposition to be associated with disruption in sleep quality even in the absence of cognitive impairment3. Furthermore, recent evidence supports a role Brendan P. Lucey and David M. Holtzman are in the Department of Neurology, Hope Center for Neurological Disorders, Knight Alzheimer’s Disease Research Center, Washington University, St. Louis, Missouri, USA. e-mail: [email protected]
for sleep in the development of AD, at least in part by influencing Aβ. Aβ fluctuates diurnally: soluble Aβ levels are higher during wakefulness and lower during sleep4,5. Sleep deprivation accelerates Aβ deposition in APP transgenic mice4, whereas orexin deficiency, which increases sleep, decreases it6. In addition, amyloid deposition disrupts sleep in APP transgenic mice7. The relationship between sleep and Aβ deposition has thus been proposed to be bidirectional: sleep disruption leads to protein deposits and protein deposits result in sleep disturbance8. The relationship among Aβ deposition, other aspects of AD pathology, sleep and memory impairment are not well defined in humans. In this issue of Nature Neuroscience, Mander et al.9 report a link between brain Aβ deposition, sleep and memory dysfunction. They hypothesize that Aβ accumulation in the medial prefrontal cortex (mPFC) is associated with diminished slow-wave activity (SWA) during non-REM (NREM) sleep that further correlates with the extent of impaired overnight hippocampus-dependent memory consolidation in older adults. Previous work from this group has shown that mPFC atrophy is associated with reduced NREM SWA
nature neuroscience volume 18 | number 7 | july 2015
and that this association correlates with overnight memory retention10. To further investigate whether amyloid deposition rather than brain atrophy has a similar effect, the authors recruited 26 cognitively normal older adults who underwent positron emission tomography imaging with Pittsburgh compound B to determine the amount of fibrillar Aβ deposited in the brain. To assess memory function, all participants trained on a set of word pairs in the evening. Then sleep was monitored overnight with polysomnography to assess different sleep stages, such as NREM sleep, and to obtain electroencephalography (EEG) for power analysis. Participants took the word-pair test again in the morning during a functional MRI scanning session. The authors found that higher amyloid burden in the mPFC correlated with decreased NREM SWA in this brain region, but not with higher frequencies of EEG activity or with decreased NREM SWA in other regions. This decrease in mPFC NREM SWA further correlated with worse overnight memory retention, even after controlling for age and sex. The authors then sought to determine the interaction of these factors using path analysis. Three models were constructed to determine the nature of 933
