NEURODEVELOPMENT

Live imaging of neurogenesis in theadult mouse hippocampusGregor-Alexander Pilz,1* Sara Bottes,1* Marion Betizeau,1,2 David J. Jörg,3,4

Stefano Carta,1,6 Benjamin D. Simons,3,4,5 Fritjof Helmchen,6 Sebastian Jessberger1†

Neural stem and progenitor cells (NSPCs) generate neurons throughout life in themammalian hippocampus. We used chronic in vivo imaging and followed geneticallylabeled individual NSPCs and their progeny in the mouse hippocampus for up to 2months. We show that NSPCs targeted by the endogenous Achaete-scute homolog1 (Ascl1) promoter undergo limited rounds of symmetric and asymmetric divisions,eliciting a burst of neurogenic activity, after which they are lost. Further, our datareveal unexpected asymmetric divisions of nonradial glia-like NSPCs. Cell fates ofAscl1-labeled lineages suggest a developmental-like program involving a sequentialtransition from a proliferative to a neurogenic phase. By providing a comprehensivedescription of lineage relationships, from dividing NSPCs to newborn neurons integratinginto the hippocampal circuitry, our data offer insight into how NSPCs support life-longhippocampal neurogenesis.

The hippocampus requires new neurons inthe dentate gyrus (DG) throughout life forlearning andmemory (1). Failing or alteredhippocampal neurogenesis has been impli-cated in a variety of diseases such as major

depression and age-related cognitive decline (2, 3).On the basis of thymidine analog labeling, in vivolineage tracing, and cell ablation studies, it hasbeen proposed that radial glia-like (R) progenitorcells are the bona fide stem cells of the adult DG(4–9). According to the prevailing model of adulthippocampalneurogenesis,R cells self-renew—heredefined as generating a daughter cell with equiv-alent molecular characteristics and potency—andgive rise to proliferative nonradial glia-like (NR)cells that divide symmetrically to generate gran-ule cells (3). However, the self-renewal capacityand lineage relationships of R cells remain con-troversial owing to the lack of longitudinal obser-vations of individual R cells and their progenywithin their niche (7, 8). Similarly to previousimaging approaches that have probed the dy-namics of somatic stem cell behavior in the non-mammalian nervous system and other stem cellniches (10–17), we used chronic in vivo imagingto track the fate of individual R cells over timewithin the adult DG.

Chronic imaging of individual brainstem cells

To label hippocampal R cells, we used mice ex-pressing a tamoxifen (Tam)–regulable Cre re-combinase under the control of the endogenousAchaete-scute homolog 1 (Ascl1) promoter crossedwith a tdTomato reporter mouse line (Ascl1-tdTomato mice) (18). Ascl1-expressing cells con-stitute an essential population of neural stemand progenitor cells (NSPCs) in the adult DG(18–20). Adult Ascl1-tdTomato mice were im-planted with a cortical window that left thehippocampal formation intact and allowed fortwo-photon imaging (Fig. 1A and fig. S1A) (21).A single Tam injection induced sparse labelingof Ascl1-expressing cells that were classified asR or NR cells on the basis of morphologicalfeatures and marker expression (Fig. 1, B andC; fig. S2; and movie S1). Only R cells wereanalyzed as a starting population. Individualclones were imaged every ~12 to 24 hours (un-less otherwise indicated) and followed for upto 2months (Fig. 1, D and E, and fig. S3). Imagedclones (n = 63) were characterized on the basisof behavioral andmorphological criteria (methods,fig. S2, movie S2, and table S1), allowing for theconstruction of individual lineage trees (Fig. 1E,fig. S3, and movie S3). After imaging, the finalfate of progeny was confirmed using immuno-histochemistry (Fig. 1, E and F, and fig. S4).In agreement with previous static clonal lineage

tracing experiments, we found that in 8- to 9-week-old Ascl1-tdTomato mice, 67% (42/63) of R cellsentered the cell cycle and became active duringthe time course of imaging. Of these active Rcells, 88% (37/42) dividedwithin the first 20 daysand, as a population, gave rise to both neuronaland glial daughter cells (Fig. 1, D to F, and figs. S3and S4, A to D) (19).Once activated, cycling R cells divided 2.3 ± 0.1

times and persisted for 9.6 ± 1.4 days on average(Fig. 2A and fig. S3). We did not find Ascl1-

targeted R cells that generated neuronal progenyand returned to long-term (>4weeks) quiescencewithin the 2-month observational period in anyanalyzed clones (Fig. 2A and figs. S3 and S4E).This suggests that, once activated, Ascl1-targetedR cells do not reenter long-term quiescence butgenerate a burst of neurogenic activity beforecommitting to terminal neuronal differentiationand loss (Fig. 2A and figs. S3 and S4, E to H). Theaverage clone size derived from active R cells was4.8 ± 0.5 cells (Fig. 2B and fig. S3). We found noevidence for terminal differentiation of R cellsinto astrocytes, which had been suggested fornestin-expressing NSPCs after several rounds ofcell division (8) (figs. S3 and S4H). Thus, the self-renewal capacity of Ascl1-targeted R cells is tem-porally limited; this finding is similar to previousresults obtained using population-based staticanalysis of nestin-labeled NSPCs (8).

Cell division history is associatedwith cell fate

We then considered the fate behavior of ac-tivated R cells and their progeny. Previously, ithas been proposed that the predominant modeof R cell division is asymmetric (6–8, 22). How-ever, without access to continuous in vivo celltracking, evidence for asymmetric fate has beenindirect. Morphological analyses of cell bodyand radial glia-like processes of R cells beforeand after cell division revealed that the mor-phology of R cells remained stable (Fig. 2, C andD, and fig. S5A), providing direct evidence forasymmetric cell divisions (6–8, 22). Althoughthe majority of observed first cell divisions wereasymmetric, generating a R cell and a NR cell(79.3%; Fig. 2, E and F), we found that 13.8% offirst R cell divisions expanded the R cell poolthrough symmetric divisions (Fig. 2, E and G),mirroring the behavior found in static clonalstudies of R cells targeted by a nestin promoter(7). With the further identification of direct neu-rogenic cell divisions of R cells (Fig. 2E and figs.S3 and S5, B and C), all three modes of divisionwere observed, reminiscent of the fate behaviordescribed for neural progenitors in the develop-ing neocortex (23, 24). The majority (70.6%) ofall R cell divisions led to the generation of NRcells that were identified on the basis of theirlack of a radial process, their ability to enter thecell cycle, and their capacity to generate neu-ronal progeny at later stages during the imagingperiod (Fig. 2, E and F). In contrast to previouslysuggested models (3), we found not only sym-metric, neurogenic cell divisions of NR cells butalso a substantial fraction of asymmetric celldivisions (24.2% of all NR divisions), yieldingone renewed NR cell and one neuronal daugh-ter cell (Fig. 2, H to J) (9). NR cells underwentas many as six rounds of cell division (with anaverage of 2.9 ± 0.2 divisions); thus, NR cells area major source of clonal expansion (Fig. 2, H toJ, and fig. S3).Chronic imaging also allowed us to analyze

whether division times (TD) are correlated withprevious cell divisions or within clonally re-lated lineages. We analyzed the TD of R and NR

RESEARCH

Pilz et al., Science 359, 658–662 (2018) 9 February 2018 1 of 5

1Laboratory of Neural Plasticity, Faculties of Medicine andScience, Brain Research Institute, University of Zurich, 8057Zurich, Switzerland. 2Institute of Neuroinformatics, Universityof Zurich and Swiss Federal Institute of Technology (ETH)Zurich, 8057 Zurich, Switzerland. 3Cavendish Laboratory,Department of Physics, University of Cambridge, CambridgeCB3 0HE, UK. 4Wellcome Trust/Cancer Research UK GurdonInstitute, University of Cambridge, Cambridge CB2 1QN,UK. 5Wellcome Trust–Medical Research Council CambridgeStem Cell Institute, University of Cambridge, CambridgeCB2 1QR, UK. 6Laboratory of Neural Circuit Dynamics,Faculties of Medicine and Science, Brain Research Institute,University of Zurich, 8057 Zurich, Switzerland.*These authors contributed equally to this work.†Corresponding author. Email: jessberger@hifo.uzh.ch

on Septem

ber 22, 2020

http://science.sciencemag.org/

Dow

nloaded from

divisions and found that, once the cells wereactivated, TD remained constant (Fig. 2K). TDtimes of sister cells were correlated among R celldaughters, whereas no correlation was detectedfor daughters of NR cells (Fig. 2, L and M), sug-gesting segregation of cellular features selectivelyin R cells that determine the daughter cells’ re-entry into cell division.

Asymmetric segregation of cell deathwithin clonal lineages

Not all newborn cells survive and become stablyintegrated into theDGcircuitry. Two critical periodsof cell death have been described previously: anearly phase of cell deathwithin the first days aftercell birth and a later phase of neuronal selectionthat is activity-dependent and occurs about 1 to3 weeks after new neurons are born (25–27). Con-sistent with previous reports, we found an aver-age frequency of cell death among the progenyper lineage of 59.6% and identified two waves

of cell death occurring around 1 to 4 and 13 to18 days after birth, respectively (Fig. 3, A and B)(25–27). We identified interclonal variability,with some clones showing no cell death, whereasother clones completely disappeared over time(Fig. 3A). Furthermore, we found differences re-garding the susceptibility to cell death amongindividual sublineages when analyzing levels ofearly cell death (until 7 days after cell birth; Fig. 3,C to F), suggesting that cell death is not evenlydistributed among progeny. The underlying causefor the observed subtree-associated variabilityof cell death remains unknown, but itmay involveintrinsicmechanisms—for example, retrotransposon-associated genetic alterations or unequal seg-regation of aging factors (28, 29)—rather thanenvironmental, niche-dependent factors, suchas growth factor availability (3). Supporting thisinterpretation, we found that surviving cellscan lie interspersed among death-prone cells(Fig. 3G).

Developmental-like program describescell fate behavior

Inspection of reconstructed R cell lineages re-vealed a wide variability in neurogenic potentialand fate outcomes (fig. S3). After induction,some R cells differentiated early, giving rise toone or two short-lived neurons, whereas othersgave rise to more than 10 surviving neurons (Fig.2B and figs. S3 and S4F). Despite this variability,some features of R cell fate behavior were con-served among all lineage trees. In particular, wefoundno instanceof anasymmetricalRcell divisionbeing followed by a symmetric R cell duplication(Fig. 2E and fig. S3), suggesting either that, onceactivated, R cells move sequentially from a sym-metrically to an asymmetrically dividing phase, orthat the proliferative potential of R cells becomesprogressively exhausted so that symmetrical dupli-cative divisions become increasingly scarce. In con-trast, variability in the output of dividing NR cellswas indicative of a conserved pattern of stochastic

Pilz et al., Science 359, 658–662 (2018) 9 February 2018 2 of 5

Fig. 1. Chronic in vivo imaging ofneurogenesis in the adult DG.(A) Scheme illustrating the experi-mental approach allowing for chronicin vivo imaging of NSPCs in the adultDG of Ascl1-tdTomato mice. (B) Rep-resentative in vivo images of R and NRcells at 2 days post-induction (dpi).(C) Immunostained images showingSox2-positive (green), Ascl1-tdTomato–labeled (red) R cells with GFAP(glial fibrillary acidic protein)–positive(white) radial processes and NR cells(Sox2-positive and GFAP-negative)in Ascl1-tdTomato mice at 2 dpi.(D) Selected imaging time points fortwo R cells (respectively indicatedwith open and filled arrowheads) overthe course of 2 months, showingthe emergence of two neuronal clones.Time points after Tam injection areindicated in each panel. Shown arecollapsed z-stacks.The clonal expan-sion of individual R cell progeny andsubsequent neuronal maturation canbe seen. (E) Lineage tree deducedfrom tracking one R cell [open arrow-head in (D)] and its progeny. Identi-fied cell types are color-coded, andlineage transitions are depicteddependingon their certainty (methods).Each circle in the lineage tree repre-sents an imaging time point.The y axisshows the duration of the imaging.(F) Post hoc immunohistochemicalanalyses of the clone shown in (D)(boxed area, day 59) confirm neuronalprogeny with newborn cells positivefor Prox1 (green) and negative forSox2 (white).The horizontal view of theDG corresponds to the view obtainedduring in vivo imaging. Scale bars,20 mm [(B) and (C)] and 50 mm [(D)and (F)]. d, days; GCL, granule cell layer.

RESEARCH | RESEARCH ARTICLEon S

eptember 22, 2020

http://science.sciencem

ag.org/D

ownloaded from

fate, with NR cells choosing between symmetricduplication, asymmetric division, and symmetricdifferentiation with probabilities independent ofthe cell generation (Fig. 2H and fig. S3).Thus, on the basis of these observations, we

considered the quantitative fate behavior of R cells.Given the sequential pattern of symmetric andasymmetric divisions, we questioned whetherR cells might be following a developmental-likeprogram, switching irreversibly from a phase of

proliferative (symmetrical) divisions to a phaseof neurogenic (asymmetrical) divisions, as ob-served during cortical development (Fig. 4A,fig. S6A, and table S2) (30). Using a statisticalmodeling approach, we quantitatively assessedthe viability of this hypothesis by fitting thelengths of putative proliferative and neurogenicphases against average clonal properties (methods).This simple developmental-like paradigm yieldedpredictions of the proliferative output, the cell

fate distributions, and the average clonal com-position over time that were in agreement withthe observed data within the theoretically pre-dicted variability (Fig. 4, B to D, and methods).As a consistency check, we also assessed whetherthe observed sequential fate pattern could rep-resent the chance outcome of stochastic fate be-havior, with R cells becoming progressivelybiased away from self-renewal toward differ-entiation over time (figs. S6 and S7). However,

Pilz et al., Science 359, 658–662 (2018) 9 February 2018 3 of 5

Fig. 2. The mode of NSPC division isassociated with individual cell divisionhistory. (A) Self-renewal duration (timebetween first and last division in eachlineage) of R cells (9.6 ± 1.3 days; n =39). (B) Distribution of the final numberof cells per active clone (n = 42lineages). Open circles represent indi-vidual clones. (C) Chronic in vivo imag-ing before and after cell divisionillustrates asymmetric cell division ofR cells. A large overlap is evident in thecellular morphology before (red arrow-head and red outline) and after (greenarrowhead and green outline) R celldivision. The black arrowhead points atthe asymmetrically generated daughtercell. (D) A morphometric index (includingcircularity and process length; detailsare given in the methods) shows littledeviation in cell morphology before andafter cell division (9.6 ± 1.8%; n = 9).(E) Heat map representing the frequenciesof modes of division of R cells (all divi-sions and division rounds 1 to 3; n = 68divisions total). The division modechanges from predominantly asymmetric(division 1) to a more symmetric differ-entiating division (divisions 2 and 3).N, neuron; A, astrocyte. (F) Example ofan asymmetric division of an R cell(lineage 40; fig. S3). (G) Example of asymmetric division of an R cell (lineage13). Mother cells are indicated with arrow-heads and daughter cells with arrows.(H) Heat map representing the frequenciesof cell division modes of NR cells (alldivisions and division rounds 1, 3, and 5; n =153 divisions). (I) Example of a symmetricNR cell division (lineage 1). (J) Exampleof an asymmetric NR cell division (lineage3). The NR daughter continues to divide.Mother and daughter cells are indicated asin (F) and (G). (K) Cell division time (TD)of R and NR cells for different divisions.(L) The times until next cell division ofsister cells originating from a single Rmother cell are correlated (Pearson’s r =0.77; *P < 0.00001; n = 22 pairs). (M) TheTD of sister cells originating from a singleNR mother are not correlated (Pearson’sr = 0.44; P = 0.08; n = 16 pairs). Eachplus sign represents a pair of sister cells.Red lines, means; error bars, SEM. Scalebars, 20 mm [(C), (F), (G), (I), and (J)].

RESEARCH | RESEARCH ARTICLEon S

eptember 22, 2020

http://science.sciencem

ag.org/D

ownloaded from

Pilz et al., Science 359, 658–662 (2018) 9 February 2018 4 of 5

Fig. 3. Chronic in vivo imaging reveals variablesusceptibility to cell death. (A) The frequency ofcell death in all chronically imaged lineages rangesfrom 0 to 100% (mean = 59.6%, n = 42; error bars,SEM). (B) Time point of cell death after last celldivision.Two peaks of cell death occur before andafter 7 days, with almost no cell death occurring>20 days after birth (n = 242). (C) Pictogramdepicting the comparison of cell death frequenciesin subtrees 1 and 2, derived after the initial R celldivision, plus subsequent progeny within that subtree(Div2plus). (D) Difference in cell death frequenciesin subtrees 1 and 2 in comparison with cell deathfrequency in the whole lineage [lineages with >25%difference in cell death rate are colored red; only earlycell death (until day 7 after birth) was included].(E) Pictogram depicting the comparison of cell deathfrequencies in the Div2plus subtrees and the sub-trees derived from the second division after initialR cell division plus subsequent progeny within thosesubtrees (Div3plus). (F) Cell death frequencies aremore asymmetrically distributed among Div3plus thanamong Div2plus sublineages, as demonstrated by thehigher weighted standard deviation to the Div2plussubtree death frequency [Div2plus difference, 13.6 ± 1.6(n = 34); Div3plus difference, 27 ± 2.2 (n = 33); *P <0.0001,Wilcoxon rank sum test].Weighted standarddeviationwasused toaccount fordifferences in subtreesizes within each clone. Red lines, means; error bars,SEM. (G) Locations of surviving newborn neurons (bluecircles) relative to cells that underwent cell death (redcircles).The observed times until death (calculatedfrom the birth of the individual cell) for dying cells are2.5days (cells 1and2 inexample 1)and13,3, and4days(cell 1, 2, and 3 in example 2). Spatial localizations ofsurviving and dying cells overlap. Scale bar, 100 mm.

Fig. 4. Modeling-based analysis suggests adevelopmental-like program for R cell fatebehavior. (A) In the model paradigm, R cellsfollow a defined programcomprisinga proliferativephase (duplications) that switches irreversibly intoa neurogenic phase (asymmetric divisions andterminal differentiation). (B) Total number ofnewborn cells in successive time intervals of5 days after induction from experiments (darkdata set) and from simulations (light data set) forR and NR cells and neurons (N) (methods). Cellcounts were pooled over 55 lineage trees withnonquiescent R cells and averaged over500 realizations for each lineage tree, resulting in atotal of 27,500 simulations (model parametersare given in table S2). Error bars indicate therange within which 95% of the simulation resultsfall. (C) Relative frequencies of different cell fates insuccessive time intervals of 5 days after inductionfrom experiments and simulations, as in (B).Later time intervals (20 to 50 days) with smallnumbers of events have been pooled together. SD,symmetric differentiating divisions; A, asymmetricdivisions; SR, symmetric self-renewing divisions;D, cell death. (D) Average clone content as afunction of time from experiments (dots) and fromsimulations (lines) for different indicated cell types.Shaded areas indicate the regions within which95% of the simulation results fall.

RESEARCH | RESEARCH ARTICLEon S

eptember 22, 2020

http://science.sciencem

ag.org/D

ownloaded from

given the observed cell fate frequencies and thenumber of observed lineage trees, we estimatedthe chance for such an outcome to be only 2.4%(methods). Thus, we conclude that a model inwhich the sporadic entry of R cells into the cellcycle activates a developmental-like program offate, leading to a burst of neurogenic activity,provides the most plausible explanation of thelineage data.In this study, we used chronic imaging of in-

dividualRcells and theirprogeny tocharacterizethecellulardynamicsunderlyingadulthippocampalneurogenesis. Our data show that, after activation,Ascl1-targetedR cells enter a developmental-likeprogram, eliciting a burst of neurogenic activity.However, self-renewal is temporally limited: Wedid not observe repeated shuttling between quies-cenceandproliferation,leadingtoalossofactivatedRcells.Thesefindingsdonotruleoutthepreviouslyreportedpresenceof stemcells in themammalianDG that shuttle back and forth between quies-cence and activity, dividing for extended periods(7,31,32). Stemcell heterogeneityhasbeenpostu-lated, and theAscl1-targeted population analyzedheremaynot includeall subtypes thatare capableof generating neuronal progeny in the adult DG(33–35). Previous data suggested that about 10to 15% of all granule cells are adult-generatedin the mouse hippocampus. This indicates thatadult neural stem cells generate 30,000 to 45,000granule cells during the entire life span (5, 36, 37),which, as a fraction of the total neuronal popula-tion, appears to be lower in the rodent than in thehuman DG (38). We found that, once activated,individual Ascl1-targeted R cells generated 4.8neurons on average. On the basis of previousestimates that the DG contains ~10,000 R cells in2-month-old mice (8), the total number of cellsthat can be generated by Ascl1-targeted cells withthe principles of clonal expansion described here

(~45,000) appears to be sufficient to explain asubstantial part of hippocampal neurogenesis.Our in vivo imaging results elucidate the cellulardynamics of physiological adult neurogenesisand form the basis to understand the molecularmechanisms governing the steps from dividingstem cells to newborn neurons.

REFERENCES AND NOTES

1. C. Anacker, R. Hen, Nat. Rev. Neurosci. 18, 335–346(2017).

2. S. Yun, R. P. Reynolds, I. Masiulis, A. J. Eisch, Nat. Med. 22,1239–1247 (2016).

3. J. T. Gonçalves, S. T. Schafer, F. H. Gage, Cell 167, 897–914(2016).

4. D. C. Lagace et al., J. Neurosci. 27, 12623–12629 (2007).5. I. Imayoshi et al., Nat. Neurosci. 11, 1153–1161 (2008).6. B. Seri, J. M. García-Verdugo, B. S. McEwen, A. Alvarez-Buylla,

J. Neurosci. 21, 7153–7160 (2001).7. M. A. Bonaguidi et al., Cell 145, 1142–1155 (2011).8. J. M. Encinas et al., Cell Stem Cell 8, 566–579 (2011).9. S. Lugert et al., Cell Stem Cell 6, 445–456 (2010).10. J. S. Barbosa et al., Science 348, 789–793 (2015).11. D. B. Gurevich et al., Science 353, aad9969 (2016).12. P. Rompolas, K. R. Mesa, V. Greco, Nature 502, 513–518

(2013).13. P. Rompolas et al., Nature 487, 496–499 (2012).14. P. Rompolas et al., Science 352, 1471–1474 (2016).15. C. Lo Celso et al., Nature 457, 92–96 (2009).16. Y. Xie et al., Nature 457, 97–101 (2009).17. L. Ritsma et al., Nature 507, 362–365 (2014).18. E. J. Kim, J. L. Ables, L. K. Dickel, A. J. Eisch, J. E. Johnson,

PLOS ONE 6, e18472 (2011).19. G. J. Sun et al., Proc. Natl. Acad. Sci. U.S.A. 112, 9484–9489

(2015).20. J. Andersen et al., Neuron 83, 1085–1097 (2014).21. G. A. Pilz et al., J. Neurosci. 36, 7407–7414 (2016).22. H. Suh et al., Cell Stem Cell 1, 515–528 (2007).23. J. H. Lui, D. V. Hansen, A. R. Kriegstein, Cell 146, 18–36

(2011).24. M. Götz, W. B. Huttner, Nat. Rev. Mol. Cell Biol. 6, 777–788

(2005).25. G. Kempermann, D. Gast, G. Kronenberg, M. Yamaguchi,

F. H. Gage, Development 130, 391–399 (2003).26. A. Sierra et al., Cell Stem Cell 7, 483–495 (2010).27. A. Tashiro, V. M. Sandler, N. Toni, C. Zhao, F. H. Gage, Nature

442, 929–933 (2006).

28. D. L. Moore, G. A. Pilz, M. J. Araúzo-Bravo, Y. Barral,S. Jessberger, Science 349, 1334–1338 (2015).

29. A. R. Muotri et al., Nature 435, 903–910 (2005).30. P. Gao et al., Cell 159, 775–788 (2014).31. G. Kempermann, Cell 145, 1009–1011 (2011).32. N. Urbán et al., Science 353, 292–295 (2016).33. M. A. Bonaguidi, J. Song, G. L. Ming, H. Song, Curr. Opin.

Neurobiol. 22, 754–761 (2012).34. Z. Chaker, P. Codega, F. Doetsch, Wiley Interdiscip. Rev. Dev.

Biol. 5, 640–658 (2016).35. N. A. DeCarolis et al., Hippocampus 23, 708–719 (2013).36. G. Kempermann, H. G. Kuhn, F. H. Gage, Proc. Natl. Acad.

Sci. U.S.A. 94, 10409–10414 (1997).37. J. Ninkovic, T. Mori, M. Götz, J. Neurosci. 27, 10906–10911

(2007).38. K. L. Spalding et al., Cell 153, 1219–1227 (2013).

ACKNOWLEDGMENTS

We thank S. April for help with analyzing R cell morphology andD. L. Moore and D. C. Lie for comments on the manuscript.Funding: This work was supported by the European ResearchCouncil (to S.J. and F.H.), the Swiss National Science Foundation(BSCGI0_157859 to S.J.), the Zurich Neuroscience Center, andthe Wellcome Trust (098357/Z/12/Z to B.D.S.). G.A.P. wassupported by a European Molecular Biology Organization Long-Term Fellowship. M.B. was supported by a SystemsX transitionpostdoctoral fellowship. Author contributions: G.A.P. developedthe imaging approach, performed imaging, analyzed data, andcowrote the manuscript. S.B. performed imaging, analyzed data,and revised the manuscript. S.C. codeveloped the imagingapproach and performed imaging. M.B. analyzed data and revisedthe manuscript. D.J.J. and B.D.S. contributed to the concept,performed theoretical modeling, and cowrote the manuscript.F.H. contributed to the concept and revised the manuscript.S.J. developed the concept and wrote the manuscript. Competinginterests: None declared. Data and materials availability: Thedata reported are presented in the main paper and thesupplementary materials.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/359/6376/658/suppl/DC1Materials and MethodsFigs. S1 to S7Tables S1 and S2References (39–54)Movies S1 to S3

27 July 2017; accepted 13 December 201710.1126/science.aao5056

Pilz et al., Science 359, 658–662 (2018) 9 February 2018 5 of 5

RESEARCH | RESEARCH ARTICLEon S

eptember 22, 2020

http://science.sciencem

ag.org/D

ownloaded from

Live imaging of neurogenesis in the adult mouse hippocampus

Sebastian JessbergerGregor-Alexander Pilz, Sara Bottes, Marion Betizeau, David J. Jörg, Stefano Carta, Benjamin D. Simons, Fritjof Helmchen and

DOI: 10.1126/science.aao5056 (6376), 658-662.359Science 

, this issue p. 658; see also p. 639Sciencethe dentate gyrus.Perspective by Götz). The results revealed the developmental progression as progenitor cells gave rise to mature cells oflabeled individual progenitor cells in the mouse hippocampus and watched them in situ for the next 2 months (see the

et al.Addition of new neurons to the adult brain is key to the hippocampal functions of learning and memory. Pilz A window on hippocampal neurogenesis

ARTICLE TOOLS http://science.sciencemag.org/content/359/6376/658

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2018/02/07/359.6376.658.DC1

CONTENTRELATED http://science.sciencemag.org/content/sci/359/6376/639.full

REFERENCES

http://science.sciencemag.org/content/359/6376/658#BIBLThis article cites 49 articles, 12 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on Septem

ber 22, 2020

http://science.sciencemag.org/

Dow

nloaded from