LARGE-SCALE BIOLOGY ARTICLE

Genomic Analysis of the DNA ReplicationTiming Program during Mitotic S Phase in Maize(Zea mays) Root TipsOPEN

Emily E. Wear,a,1 Jawon Song,b Gregory J. Zynda,b Chantal LeBlanc,c,2 Tae-Jin Lee,a,3 Leigh Mickelson-Young,a

Lorenzo Concia,a Patrick Mulvaney,a Eric S. Szymanski,a,4 George C. Allen,d Robert A. Martienssen,c

MatthewW. Vaughn,b Linda Hanley-Bowdoin,a and William F. Thompsona

a Department of Plant and Microbial Biology, North Carolina State University, Raleigh, North Carolina 27695b Texas Advanced Computing Center, University of Texas, Austin, Texas 78758cCold Spring Harbor Laboratory, Cold Spring Harbor, New York 11724dDepartment of Horticultural Science, North Carolina State University, Raleigh, North Carolina 27695

ORCID IDs: 0000-0003-2002-6980 (E.E.W.); 0000-0002-9916-3816 (J.S.); 0000-0003-4992-9614 (G.J.Z.); 0000-0002-2848-4970(L.M.-Y.); 0000-0002-7401-7214 (L.C.); 0000-0003-2582-6573 (E.S.S.); 0000-0002-1384-4283 (M.W.V.); 0000-0001-7999-8595(L.H.-B.)

All plants and animals must replicate their DNA, using a regulated process to ensure that their genomes are completely andaccurately replicated. DNA replication timing programs have been extensively studied in yeast and animal systems, but muchless is known about the replication programs of plants. We report a novel adaptation of the “Repli-seq” assay for use in intactroot tips of maize (Zea mays) that includes several different cell lineages and present whole-genome replication timing profilesfrom cells in early, mid, and late S phase of the mitotic cell cycle. Maize root tips have a complex replication timing program,including regions of distinct early, mid, and late S replication that each constitute between 20 and 24% of the genome, as well asother loci corresponding to ;32% of the genome that exhibit replication activity in two different time windows. Analyses ofgenomic, transcriptional, and chromatin features of the euchromatic portion of the maize genome provide evidence fora gradient of early replicating, open chromatin that transitions gradually to less open and less transcriptionally active chromatinreplicating in mid S phase. Our genomic level analysis also demonstrated that the centromere core replicates in mid S, beforeheavily compacted classical heterochromatin, including pericentromeres and knobs, which replicate during late S phase.

INTRODUCTION

DNA replication enables a cell to faithfully transmit genetic ma-terial to daughter cells while maintaining genome integrity. Duringeach S phase, theremust be a complete and accurate duplicationof the genome, which must occur in a regulated and reproducibleway from one cell cycle to the next (Klein and Gilbert, 2016). Toaccomplish this task,higher eukaryotesorganize theprocessbothtemporally and spatially so that replication initiates atmultiple loci,or origins, distributed throughout the genome, with different ori-gins becoming active at different times during S phase (Masai

et al., 2010). In most higher eukaryotes, there is an associationbetween early replication and euchromatic, transcriptionallyactive chromatin (Hatton et al., 1988; Hiratani et al., 2008; Leeet al., 2010;Rybaet al., 2010), though there are classesof genesthat have a much weaker association with replication time andare developmentally regulated in humans and mice (Hirataniet al., 2008; Rivera-Mulia et al., 2015). Conversely, there isalso a long established association between late replicationand classical heterochromatin (Lima-de-Faria and Jaworska,1968; Pryor et al., 1980), although there are a few exceptions(Kim et al., 2003). However, the intervening time between whatis considered “early” and “late” has not been characterizedthoroughly by whole-genome replication timing analyses, andreplication times in this portion of S phase are only inferred fromthe ratio of signals obtained from early versus late in manystudies. Some information is available about mid replicatingchromatin from cytological studies. For example, in mamma-lian cell lines, mid replicating loci are spatially localized to theperinuclear and perinucleolar edges, while early replicationoccurs in dispersed, punctate foci (Dimitrova and Berezney,2002; Panning and Gilbert, 2005; Zink, 2006). Such spatio-temporal differences are less pronounced in plants, in whichboth early and mid replicating loci have been reported to be

1Address correspondence to emily_wear@ncsu.edu.2 Current address: Department of Molecular, Cellular, and DevelopmentalBiology, Yale University, New Haven, CT 06511.3 Current address: Syngenta Crop Protection, Research Triangle Park,NC 27709.4 Current address: Department of Biochemistry, Duke University, Dur-ham, NC 27710.The author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: William F. Thompson(wftb@ncsu.edu).OPENArticles can be viewed without a subscription.www.plantcell.org/cgi/doi/10.1105/tpc.17.00037

The Plant Cell, Vol. 29: 2126–2149, September 2017, www.plantcell.org ã 2017 ASPB.

dispersed throughout the nucleoplasm (Samaniego et al.,2002; Bass et al., 2015).

Replication inmidSphase has also been indirectly discussed inthe context of timing transition regions (TTRs), which connectadjoining earlier and later constant timing regions in mammals(Hiratani et al., 2008; Desprat et al., 2009; Ryba et al., 2010). Somehave hypothesized that these TTRs consist of unidirectionalreplication forks spreading from origins active earlier in S phase(Hiratani et al., 2008; Ryba et al., 2010), while others have arguedthat such regions contain origins activated in a “cascading” or“domino” pattern by earlier firing origins nearby (Guilbaud et al.,2011). There has been much less debate over the patterns ofactivation of origins firing within the constant timing region do-mains, as these are thought to activate roughly synchronously inclusters (Jackson and Pombo, 1998; Blow et al., 2011; Bouloset al., 2015; Klein and Gilbert, 2016).

Most studies in mammalian cells agree that much of the rep-lication occurs in large coordinated domains (0.2–2 Mb) thatpresumably contain multiple origins (Farkash-Amar et al., 2008;Guilbaud et al., 2011; Hyrien, 2016; Klein and Gilbert, 2016). Thesize of replication domains in mammals is further supported byresults from chromatin capture experiments that have definedtopologically associated domains. Topologically associated do-mains align with the early boundaries of TTRs that delineatereplication timingdomains (Popeet al., 2014).However, in specieswith smaller genomes, such as Drosophila melanogaster andArabidopsis thaliana, replication domains are smaller, with ameandomain size in the range of;30 to 450 kb (MacAlpine et al., 2004;Schwaiger et al., 2009; Lee et al., 2010).

Although plant and animal lineages diverged 1.6 billion yearsago, the mechanisms and core machinery used to duplicate DNAhave been largely preserved (Shultz et al., 2007; DePamphilis andBell, 2010). However, there has been little attention focused onunderstanding DNA replication programs in plants (Lee et al.,2010; Costas et al., 2011a). Early literature using tritiated thymi-dine autoradiography indicated that plant-specific environmentaland developmental events can impact the number and spacing ofreplication origins (reviewed in Bryant and Aves, 2011). Recentwork revealing differences between transcriptional regulationstrategies (Hetzel et al., 2016) also illustrates how fundamentalprocesses, often assumed to be the same in all eukaryotes, maydiffer in important aspects between plants and animals.

In plants, new organs continue to form throughout an individ-ual’s lifespan, providing a unique and useful model for in-vestigating DNA replication dynamics in both space and time(Costas et al., 2011a). Early research on plant DNA replicationfocused mostly on the duration of S phase, the rate of forkmovement, the size of individual replicons, and the number ofreplicon families that initiate replication coordinately. Thesestudies were largely based on cell cycle kinetics and DNA fiberautoradiography experiments primarily using root tip meristemcells from various plant species (Van’t Hof, 1976; Van’t Hof et al.,1978a, 1978b; Van’t Hof and Bjerknes, 1981; Kidd et al., 1987).More recently, studies also characterized plant homologs ofproteins involved in the prereplicative complex (Kimura et al.,2000; Castellano et al., 2001; Ramos et al., 2001; Witmer et al.,2003; Masuda et al., 2004; Mori et al., 2005; Shultz et al., 2007).However, genomic analyses of DNA replication and replication

origins in plants have been sparse and thus far have focusedexclusively on Arabidopsis. Replication timing has been profiledfor a single chromosome (Lee et al., 2010), and efforts have beenmade to identify initiation zones or origins of replication (Lee et al.,2010; Costas et al., 2011b).Becauseof itssmall genomesizeand relativepaucityof repeats,

it is reasonable to suppose that DNA replication in Arabidopsismaybesimpler than for the larger,more complexgenomesof cropplants. Maize (Zea mays) has a moderately large (2.3 Gb), fullysequenced genome (Schnable et al., 2009). The complexity of thisgenome is similar to that of many other crop plants, with a highconcentration of repetitive sequences located in heterochromaticblocks and extensively interspersed in the euchromatic portion ofthe genome. Together with its illustrious history as a modelspecies for genetic and cytogenetic studies, these features makemaize an ideal model crop system for investigating the replicationtiming program.We developed a system (reviewed in Bass et al., 2014) to ex-

amine the spatial, temporal, and genomic patterns of DNA rep-lication in maize, using rapidly cycling cells from root tips pulse-labeled with the thymidine analog 5-ethynyl-2’-deoxyuridine(EdU). EdU offers a substantial improvement over classical rep-lication timing assays, which have typically used 5-bromo-2’-deoxyuridine (BrdU) to label and immunoprecipitate newly repli-catedDNA (Hiratani etal., 2008;Schwaigeret al., 2009;Chenetal.,2010;Hansenet al., 2010; Leeet al., 2010;Rybaet al., 2011). A keybenefit of using EdU is that a heat or acid denaturation step is notrequired for immunoprecipitation (reviewed in Darzynkiewiczet al., 2011). Hence, it is possible to visualize EdU-labeled DNAwhile maintaining native subnuclear structure (Kotogany et al.,2010; Bass et al., 2014) and to separate labeled from unlabelednuclei by flow cytometry prior to DNA isolation (Wear et al., 2016).We previously used the EdU labeling system to investigate thespatio-temporal aspects ofDNA replication in proliferating root tipcells (Bass et al., 2015). In that study, we used fluorescence mi-croscopy to show that loci replicating in early andmid S phase arein close proximity to each other but have limited spatial overlap.This observation led us to hypothesize that there are at least two“euchromatic compartments” in replicating nuclei.We envisionedthat the early replicating compartment is largely comprised ofextended, transcriptionally active, accessible chromatin, whilechromatin replicating in mid S phase includes locally compactedrepetitive blocks and silent genes (Bass et al., 2015).In this study,weextendourpreviouswork to includeafine-scale

genomic analysis of DNA replication, gene expression, andchromatin accessibility to add detailed molecular evidence to theinitial cytological observations. Our study represents the first tocharacterize a replication timing program in a crop species, andthe first whole-genome analysis in any plant species. The root tipsystem allowed us to work with intact meristems, avoiding po-tential chromosome aberrations, mutations, aneuploidization,and other genetic and chromatin-related changes that have beendocumented in both plant and animal cell cultures (Lee, 1988;Phillips et al., 1994; Serrano and Blasco, 2001; Maitra et al., 2005;Tanurdzic et al., 2008; Mayshar et al., 2010; Laurent et al., 2011).We present whole-genome profiles of replication activity in

early, mid, and late S phase and companion data sets for tran-scription, and a selection of histone marks. We found that maize

DNA Replication Timing Program in Maize 2127

root tips have a complex replication timing program, includingregions of distinct early, mid, and late S replication that eachconstitute between 20 and 24% of the genome, as well as lociaccounting for another ;32% of the genome that exhibit repli-cation activity in twodifferent timewindows, suchasearly andmidor mid and late. Analyses of genomic, transcriptional, and chro-matin features of the euchromatic portion of the maize genomeprovide evidence for a gradient of early replicating, open chro-matin that transitions gradually into less open and less tran-scriptionally active chromatin replicating in mid S phase. Ourresults also confirm previous observations in maize nuclei thatheavily compacted classical heterochromatin replicates late inS phase (Pryor et al., 1980; Bass et al., 2015).

RESULTS

Whole-Genome Profiling of DNA Replication Timing in MaizeRoot Tips

A flowchart of the experimental steps used to generate replicationtiming by sequencing or “Repli-seq” data for maize root tips isdetailed in Figure 1A.Wepulse labeled 3-d-oldB73 seedling rootsin vivowith the thymidine analog, EdU, for 20min to label DNA thatwas replicated during the pulse period. Over the 20-min labelingperiod, EdU uniformly penetrates the root and labels cells in themultiple emerging cell lineages present in themeristematic region(Figure 1B).Under thesamegrowthconditions, theaverage lengthof S phase of mitotic cells in the terminal 1-mm region of the rootwas estimated to be between 2.7 and 3.9 h (Mickelson-Younget al., 2016). Hence, a 20-min labeling period represents;10%ofthe length of S phase. The terminal 1-mm root segments (Figure1B) were excised, fixed, and frozen before isolating nuclei for flowcytometric analysis (Figure 1C; Supplemental Figure 1A). Most ofthe nuclei from the terminal 1-mm region had DNA contentsranging from 2C to 4C, characteristic of cells undergoing amitoticcell cycle (Baluska, 1990). Approximately 20% of the nuclei hadDNAcontents above4C, indicative of somecells transitioning intoa programmed endocycle (Figure 1C; Bass et al., 2014, 2015). In1-mm root tips, nuclei from endocycling cells can be excludedbased on DNA content as shown in Figure 1C and are not con-sidered further in this article.

Maize root tips yield relatively pure nuclei preparations that areamenable to flowsorting (Figure 1C; Supplemental Figure 1A).Wedeveloped a protocol for bulk nuclei isolation from root tips ex-cised from;500seedlings (Wear et al., 2016) that is amodificationof the original tissue chopping method described by Galbraithet al. (1983). We visualized EdU-labeled nuclei by conjugatingan Alexa Fluor-488 (AF-488) fluorophore to EdU using “clickchemistry” (Salic and Mitchison, 2008). The EdU-labeled nucleiwere then separated from unlabeled nuclei by flow cytometry andsorted into populations representing early, mid, and late stages ofthe mitotic S phase based on DNA content (rectangle areas inFigure 1C).We intentionally left space between the early, mid, andlate sorting gates to reduce the amount of overlap in the sortedpopulations. Unlabeled G1 nuclei were also sorted to provide anunreplicatedDNA reference for subsequent analysis. In Figure 1D,the range of DNA contents for the early, mid, and late S sorting

gates are overlaid on theDNA content histogram for total nuclei todemonstrate that nuclei inearly and lateScannotbe resolved fromnuclei in G1 and G2/mitosis on the basis of DNA content alone.Separationof thesepopulationscanonlybeachievedusinga two-color sorting strategy (e.g., 49,6-diamidino-2-phenylindole [DAPI]and EdU/AF-488) to separate labeled and unlabeled nuclei. Asmall portion of nuclei from each sorting gate was reanalyzed todetermine the sort purity (Supplemental Figure 1B). The reanalysisshowed that there was very good separation between the his-tograms of DNA content for each sorted fraction with only ;5%overlap between adjacent fractions (Supplemental Figure 1C).EdU/AF-488 labeled DNA from the sorted nuclei in early, mid,

and late S phase was immunoprecipitated (IP) using an antibodyspecific to the AF-488 moiety. In our experience, the AF-488DNA-IP is highly reproducible, both in efficiency of precipitation(see Methods) and in quality of the sequencing data obtainedwhen using it. EdU/AF-488-labeledDNA fromeachSphase stagefrom three biological replicates, aswell as unlabeledG1 referenceDNA, was sequenced on an Illumina HiSeq 2000 to generatepaired-end 100-bp reads. After quality control and trimming, weobtained over 100 million read pairs per S phase stage thatuniquely mapped to the maize B73 AGPv3 reference genome,representing 9 to 17x whole-genome coverage for each S phasesample (see Supplemental Table 1 for mapping statistics). Eventhough the maize genome is highly repetitive and transposonladen, there is sufficient sequence polymorphism to allowmappingof the majority of reads to the reference genome (Schnable et al.,2009; Gent et al., 2013). The use of paired-end sequencing alsoaided significantly in mapping reads uniquely.Wegeneratedaprofileof the replicationactivity in eachS-phase

stage across the genome using a custom computational pipelinecalledRepliscandescribed indetail byZyndaetal. (2017).The readdensitieswereaggregated into1-kbwindows, andafter observinga strong Pearson correlation of 0.8 to 0.98 between the biologicalreplicates (Supplemental Figure 2), the reads in each window ofthe replicateswere summed (Figure1F). Artificially highor very lowread coverage regions were excluded from the analysis by re-moving statistically outlying coverage. The total readnumber fromeach S phase sample and the G1 reference was then scaled tocorrect for overall sequencing depth differences, allowing com-parison of the signal in corresponding genomic intervals betweenthe different S phase stages. Read counts fromeach1-kbwindowineachS-phase stagewere thendividedby the readcounts for thecorresponding genomic interval in the unlabeled G1 data set(Figure 1G), which provided a reference in which all genomicsequences have a 2C copy number. Normalizing to G1 correctedfor different sequencing efficiencies, yielding an estimate ofreplication intensity in each 1-kb window. Finally, the data weresmoothed by Haar wavelet transform (Figure 1H) to reduce noisewithout altering peak boundaries (Percival and Walden, 2000).Figures 1E to 1H show a visual summary of these data processingsteps for the early S data in an example region of the genome. Thecorresponding data formid and late S in the same genomic regionare in Supplemental Figure 3. These analyses resulted in whole-genomeprofiles of the intensity of theDNA replication occurring in1-kb windows from cells in early, mid, or late S phase (Figure 2B)visualized using the Integrative Genomics Viewer (IGV; Robinsonet al., 2011).

2128 The Plant Cell

The Temporal Order of Replication on Maize Chromosomes

ThemaizeB73genomeconsistsof10metacentricchromosomes,which range in size from 150 to 301 Mb in the AGPv3 genomeassembly. Protein coding genes aremost densely represented onthe ends of the chromosome arms (Schnable et al., 2009), thoughgenes are present in all major portions of each chromosome,including the centromere (Zhao et al., 2016). Likemany other largegenomes, maize is notable for an abundance of transposableelements (TEs), which are estimated to account for ;85% of thegenome (Schnable et al., 2009).We observed several global trendsfor replication activity across the maize chromosomes: (1) thehighest intensity of early replication coincideswith the gene-dense

ends of the chromosome arms (see early S profile in Figure 2B andSupplemental Figure 4); (2) the highest intensity of late replicationcoincides with the pericentromeric centers of the chromosomes(see late S profile in Figure 2B and Supplemental Figure 4); and (3)midS replication ismore evenly dispersed along the chromosomesand only slightly higher near the middle of the chromosome arms(see mid S profile in Figure 2B and Supplemental Figure 4).A replication timing profile for the entire chromosome five is

illustrated inFigure2B,whichshowstracks forearly,midand lateSreplication along with corresponding tracks for gene and TEcoverage (Figure 2A). When we further expand our view alongchromosome five to smaller example regions (2.5 Mb in size),the finer structure of peaks and valleys of early, mid, and late

Figure 1. Experimental Approach.

(A)Workflow. Roots of 3-d-oldmaize seedlingswere pulse labeledwith EdU for 20min, after which terminal 1-mmsegmentswere harvested and fixedwithformaldehyde.(B)Amergedconfocal imageofa1-mmroot tip longitudinal sectionshowingDAPIstainedDNA (red)andEdU label innewly replicatedDNA (green). Therearemultiple emerging cell lineages present in the terminal 1 mm of the root. Bar = 100 mm.(C) Sorting. Nuclei were isolated and EdU incorporated into DNA was conjugated to a fluorescent probe (AF-488) using click chemistry. Nuclei werecounterstained with DAPI prior to sorting by flow cytometry using 355-nm (UV) and 488-nm (blue) lasers. A bivariate plot of relative DNA content (DAPIfluorescence) andEdU incorporation (AF-488fluorescence) is shown,overlaidwith thegates (black rectangles) used tosort nuclei representingearly (E),mid(M), and late (L) fractions of S phase. Unlabeled nuclei from G1 phase (G1) were also sorted to use as a reference.(D)Histogram showing relative DNA content (DAPI) for the unsorted nuclei population (black line), overlaid with the position and relative frequency of nucleithat fall in the indicated sortinggates.DNAwasextracted fromsortednuclei andEdU/AF-488-labeledDNA immunoprecipitated from theearly,mid, and latefractions with an AF-488 antibody, prior to sequencing on the Illumina HiSeq 2000 platform.(E) to (H) Summary of computational processing of Repli-seq reads.(E) and (F) The number of reads that mapped uniquely to the maize B73 AGPv3 reference genome was calculated over 1-kb windows (see Methods).(G)After normalization for sequencingdepth, replication activity was expressed as the ratio of EdU/AF-488 reads in early,mid, or late Sphase to reads fromtotal DNA from unlabeled G1 nuclei.(H) The resulting data smoothed with a Haar wavelet function. Representative data tracks from IGV are shown here for early S data, and the correspondinggenomic region is shown for early, mid, and late S data in Supplemental Figure 3. Artificial spikes in sequencing coverage (arrowheads) often correspond totandem repeat regions that have been “collapsed” in the reference assembly, and these regions are subsequently excluded. Scale: E and F, 0 to 1200 readdensity; G and H, 0 to 5.4 normalized signal ratio.

DNA Replication Timing Program in Maize 2129

Figure 2. Chromosome 5 Replication Profiles.

(A)GeneandTEcoverageweredeterminedusing themaizegenomeAGPv3annotationandareexpressedas thegeneorTEpercent coverage, respectively,in 10-kb nonoverlapping windows. For IGV visualization, the coverage values were smoothed (see Methods). Gray dashed line represents 50% coverage.(B)Replication intensity profiles for early,mid, and lateSphasecells processedandpresented asdescribed inMethods and theFigure 1 legend.Scale for allreplication intensity tracks is 0 to 4.5 normalized signal ratio.(C) Segmentation of replication timing profiles into the predominant replication time classes (RT classes; see Methods) for each 1-kb window. ReplicationtimingwasclassifiedasE (early), EM (early andmid),M (mid),ML (midand late), L (late), EL (early and late), EML (early,mid, and late), andNS (not segmented);see color chart at bottom of figure.(D)Schematic representationof chromosome5with the centromere position approximatelymarkedbasedon theCENH3binding region (Zhaoet al., 2016).Red rectangles denote the locations of the expanded panels shown below, each 2.5 Mb in size.(E) A region near the end of the short arm composed predominantly of early replication and E segments.(F)and (G)Regionsnear themiddleof theshort armcomposedofvariouscombinationsof singleandmixedRTclasssegments.Regionswitha relatively highabundance ofEMandMLsegments often showclear replication activity at both times (F)or occur in regionswhere replication activity is spreading along thechromosome as S phase proceeds (G).(H)Apericentromeric regionshowingpredominantly late replicationandLsegments.Corresponding tracks for geneandTEpercentageof coverageand thecomposite RT class segmentation are displayed in each expanded panel, as described in (A) and (C).(I) The RT segment classification color legend.

2130 The Plant Cell

replication interspersed with each other is apparent. The moredetailed patterns are consistent with, but more complex than, theglobal patterns listed above. For instance, peaks of early repli-cation activity often occur in regions with higher gene coverage(Figures 2E and 2G), and peaks of late replication activity aremostoften associated with regions of lower gene and higher TE cov-erage (Figures 2G and 2H).

Though all maize chromosomes generally display similar globalreplication timingpatterns, inspectionof thepatterns for individualchromosomes in detail revealed many interesting differences(Supplemental Figures 5 and 6). For example, chromosome four,whichhas the lowest genedensity (17genes/Mb), alsohasseveraldispersed large blocks (0.7–3 Mb) of late replication that occur inthe middle and ends of the chromosome arms (SupplementalFigure 6A). Another example of chromosome variability is chro-mosome six, which contains both a knob and the nucleolus or-ganizer region on its short arm, as seen by fluorescence in situhybridization (Albert et al., 2010), and also has an asymmetricallypositioned centromere in the reference genome assembly. Thereplication timing pattern of chromosome six reflects this asym-metrywithmore intenseearly replicationoccurringon the longarm(Supplemental Figure 6B). However, it should be noted that thereare known difficulties with portions of the assembly of chromo-some six (Bilinski et al., 2015; Gent et al., 2015; Zhao et al., 2016).As theB73 referencegenomeassembly improves in the future,wewill beable to furtherfine-tuneour understandingof the replicationtiming patterns on this chromosome.

The only similar study profiling DNA replication timing in plantsexamined Arabidopsis chromosome four (Lee et al., 2010). Thatstudy, which used a 1-kb tiling microarray and a 1-h BrdU pulse,concluded that the replication timing program was essentiallybiphasic,meaning the early andmidS replication intensity profileswere nearly identical for most of the euchromatic portion ofchromosome four, while heterochromatin replicated separately inlate S. This initial finding prompted us to assess whether thereplication program of the much larger maize genome with morerepetitive DNA follows a similar pattern or is more complex. Wefound evidence that maize does have a more complex temporalreplication program, in which the global distribution of replicationactivity inmid S is clearly different from the patterns in early or lateS (Figures 2B and 2E to 2H; Supplemental Figures 4 and 5).However, there is someoverlap of both early and late S replicationtiming patterns with mid S. Interestingly, in regions with distinctmidSpatterns, peaks ofmid replication often alternatewith peaksof either early or late replication, consistent with the idea thatreplication activity may spread from one temporally ordered“zone” or “domain” of coordinate replication into another (Figures2G and 2H). These replication zones likely contain clusters oforigins that initiate replication or “fire” somewhat synchronously,althoughnot every origin in a cluster is expected to fire in every cellor in every round of replication (Blow et al., 2011; Hyrien, 2016).

One thing immediately apparent from the genomic replicationtiming profiles is the presence of regions exhibiting heterogeneityin replication time. There are many examples of regions wheresome level of replication occurs in more than one, or even in allthree, S-phase fractions (e.g., see replication intensity profiles inFigure2F). Thisphenomenonmaybemoreobviousbecauseof theway the data are presented. For example, if “replication time” is

defined as a ratio of replication activity at two discrete times in Sphase (e.g., early versus late), the procedure imposes a singlereplication time for each locuseven though that locusmayactuallydisplay activity atmultiple times. Instead,wedisplay the data fromeach S-phase fraction separately, allowing us to observe repli-cation at multiple times. There are several potential sources ofheterogeneity in replication time, one of which is the presence ofseveral cell types in the root tip, each of which may have differentreplication timing programs.

Classifying Predominant Replication Time acrossthe Genome

To enable the maize replication timing data to be associated withother genomic and epigenomic data, we sought to identifya predominant replication time for each 1-kb window of the ge-nomewhile still retainingasmuch information aspossible from thereplication intensityprofiles.Weaccomplished thisusinga robust,custom segmentation algorithm that uses the replication timingdata to automatically define segmentation parameters, instead ofarbitrarily setting them by trial and error (Zynda et al., 2017). Thewhole-genome segmentation algorithm first defines a thresholdabovewhich signals are considered replicating and then classifiesregions containing signals above the thresholdwith respect to thepredominant replication time(s). The classification considers therelative intensity of replication signals in early, mid, and late Sphase for each 1-kb window. The largest signal value in eachwindow is always classified as replicating, and the algorithm al-lows multiple-time classifications (e.g., early and mid) when an-other signal is within 50% of the highest value.The segmentation of predominant replication time (RT) classes

is illustrated in Figure 2 for all of chromosome five (Figure 2C) andfor four 2.5-Mb example regions (Figures 2E to 2H). RT classi-fications for other chromosomes are shown in SupplementalFigure 6. The classifications include the discrete single-timeclasses, early (E), mid (M), and late (L), as well as their adjoiningmultiple-time classes, early-mid (EM) andmid-late (ML). The totalgenome coverage is similar for E (21.8%), M (23.7%), and L(20.1%), with slightly less classified as EM (17.8%) and ML(14.0%) (Figure 3A, Table 1). The segmentation algorithm alsoallows for early-late (EL) and early-mid-late (EML, pan S) classesthat together only comprise 0.5% of the genome (Figure 3A). Themultiple-time classes represent regions where replication is oc-curring strongly at more than one time. In the end, we classified;98% of the assembled maize genome into either a single ormultiple-time RT class. Given that 97.4% of the genome fell intothe fivemain classes (E, EM,M,ML, andL), thesewere the primaryfocus for further analysis.

Characterizing Predominant Replication Time Classes

After segmenting themaizegenome intopredominantRTclasses,we investigated the differences in size and number of individualsegments in each class. There are fewer overall segments cate-gorized as predominantly E or L (Figure 3B) but the median size islarger (see Figure 3C for size distribution) than the segments inother classes. The larger segments are consistent with the rep-lication intensity profiles, in which strong early or late replicating

DNA Replication Timing Program in Maize 2131

regionsoftenoccur in largeblocks (e.g., seeFigures2Eand2H).Msegments, which have the most abundant genome coverageoverall (23.7%), are more numerous but generally smaller than Eand L segments (Figures 3B and 3C), consistent with the ob-servation thatmid S replication activity is fairly evenly dispersed inlower intensity blocks across chromosomes (e.g., see Figures 2Fand 2G). Segments categorized as EM and ML are similar to M insegment size (Figure 3C) and are most often found either in thetransition regions between spreading peaks (e.g., see Figure 2G)or in regions where there is heterogeneity in replication time (e.g.,see Figure 2F). We found a generally consistent distribution of RTclasses on all 10 chromosomes, with some variability betweenchromosomes for more E or L class coverage (Figure 3D), re-flecting the unique timing patterns associated with features thatare chromosome specific. We also analyzed the percent GCcontent and found it varied only slightly between RT classes(Supplemental Figure 7).

Association of Genes and Transcriptional Abundance withReplication Time

Wenext examined the association of the different RT classes withother genomic features in maize. We already noted the high in-tensity of early replication in the gene-rich ends of chromosomearms, but E segments are also found inmanyother places, andwewanted to more fully understand the relationship between repli-cation time andall genes, not just the ones at the endsof arms.Weused the AGPv3 5b+ filtered gene set (FGS) annotation, whichcontains 39,475 putative protein coding genes with evidence forfunctionality. When we considered the total coverage in 1-kbwindows of each RT class of segments across the genome andcalculated the percentage of each RT class that overlapped theboundaries of a gene, we found that E segments overlap;48%ofthe genes (Table 1). However, only 21% of the 1-kb windows thatmake up the E segments physically overlap a gene (Figure 4A),indicating that many such regions are not associated with a cur-rently annotated gene. Other RT classes have reduced associa-tion with genes, as measured both by the number of genes andpercent of the total windows in each class that overlap a gene(Table 1, Figure 4A). We used a permutation analysis approach toassess the statistical significance of the percent overlap valuesfound between each RT segment class and genes (seeMethods).We tested the relationships in both directions, representing twodifferent hypotheses. For example, to test the significance of theenrichment of genes in RT segment classes, we randomly shuf-fled RT segments; conversely, to test the significance of the

Figure 3. Characterizing RT Class Segments.

A segmentation procedure was performed (see Methods) to identify thepredominant replication time for each 1-kb window across the genome,and adjacent windows with the same RT class were merged.

(A)Total coverageofeachRTsegmentclassacross theentiregenome.Thepercentage of the genome covered by each class is noted inside the bars.(B) The number of individual segments in each RT class.(C) Box plot of the distribution of segment sizes in each RT class. Box plotwhiskers represent 1.53 interquartile range (IQR).(D) The percentage of proportion of each RT class on individual chro-mosomes. The entire genome comprises 65.6% single-time segmentclasses of E (21.8%), M (23.7%), or L (20.1%). Another 32.3% comprisesmultiple-time segment classes, EM (17.8%), ML (14.0%), EL (0.2%), andEML (0.3%). Inall,;98%of thenucleargenomewasassigned toasegmentclass. See Figure 2I for RT class color legend.

2132 The Plant Cell

enrichmentofanRTsegmentclass ingenes,we randomlyshuffledgenes. We found that the observed enrichment for genes in boththeE andEMclasses (asterisks in Figure 4A; Supplemental Table 2)and the enrichment for E andEMsegments in genes (SupplementalTable 3) are both highly significant compared with those expectedby chance (permutation P value = 0.001).

We also asked if replication timing varies with gene density. Asameasure of local genedensity, we identifiedgenes found in eachRT class and measured the distance from their 59 and 39 ends tothe nearest genes on each side, regardless of the RT class wherethenearest genewas found.Themediannearestgenedistance forgenes in E segments is ;10 kb, with a large increase in the dis-tribution of genedistances for later replication classes (Figure 4B).For example, the median gene distance is 43 kb for M segmentsand 73 kb for L segments. Thus, E segments are found in themostgene dense regions of the chromosome. In spite of this, 79% of Ewindows do not actually overlap a gene (Table 1). However, suchregions are usually in close proximity to genes but also containother features, such as TEs.

Additionally, we assayed transcriptional activity by RNA-seqin the same terminal 1-mm root tip region used to generate theRepli-seq data and evaluated the distribution of transcriptionlevels for genes in each RT class (Figure 4C). The normalizedgene expression values, expressed as fragments per kilobase oftranscript per million mapped reads (FPKM), were calculated forFGS genes. Genes in each RT class were then ranked into fivegene expression levels, ranging from no expression (FPKM = 0)to >100 FPKM. The majority of genes (65%) are low to moder-ately expressed, but all gene expression levels that have anFPKM greater than zero are strongly represented in E segments(Figure 4D). By contrast, genes with an FPKM of zero are moreevenly distributed among the RT classes (Figure 4D). Further-more, even though there are fewer genes expressed at highlevels, the percentage of genes replicating early increased withexpression level, with the E class including 72% of the genes inthe highest expression level (FPKM > 100; Figure 4E). However,there are some exceptions of very highly expressed genes inother RT classes. Taken together, our data show that regionsundergoing early replication are enriched for genes, particularlyfor genes expressed at higher levels.

Association of Transposable Elements withReplication Time

Collectively, TEs of various kinds comprise most of the maizegenome. To assess their relationship(s) to replication time, weused the repeat region annotations from AGPv3 to calculate thetotal TE coverage in 10-kbwindows and quantify the TE coveragein segments fromeachRTclass. The10-kbwindowswere used toavoid the overwhelming predominance of 0 and 100% coveragevalues that result when 1-kb windows are used. The median TEcoverage in E segments is surprisingly high at 85%, with a steadyincrease up to 98% in L segments (Figure 5A). However, within Esegments there is a much broader distribution of the two centralquartiles for TE coverage (52–99%) relative to L segments (89–100%) (Figure 5A). TEs are traditionally thought to be associatedwithheterochromatic regionsof thegenome, so thehighpresenceof TEs in E and EM classes is interesting and warranted furtherinvestigation. Hence, we decided to examine individual familieswithin the most abundant group of TEs, the class I long terminalrepeat retrotransposons (LTR-retros).In the maize genome, the LTR-retros constitute ;75% of the

genome sequence (Baucom et al., 2009). Although there are over400LTR-retro families, the top20mostabundant familiesmakeup;70% of the genome and are comprised of four families from theRLC/copia superfamily, 12 from the RLG/gypsy superfamily, andfour from the RLX/unknown superfamily (Baucom et al., 2009).Annotations in AGPv3 were used to calculate the coverage foreach of the top 20 LTR-retro families in each RT class. Takentogether, the distribution of the top 20 LTR-retros do not differdramatically from the overall genome distribution of the RTsegment classes (compare Figures 3A to 5B), except the totalcoverage in E segments is lower than expected for a completelyeven distribution (Figure 5B). The percentage of coverage for thetop 20 LTR-retros is 59% for E segments and ranges from 79 to88% for the other RT classes (Figure 5B, Table 1). However, in-dividual families differ dramatically with respect to their distribu-tion across chromosomes (Supplemental Figure 8A;Mroczek andDawe, 2003; Lamb et al., 2007). We grouped the families ac-cording to replication time trend (“earlier,” “middle,” and “later”)andordered thembyabundance.Thesixmostabundantof the top

Table 1. Summary of Genome Composition and Features of RT Classes

Coverage (Mb) FGS

RT Class Total %a Genic %b Intergenic %b Top 20 LTR-Retro %b Gene Countc %d Median FPKMe

E 449.9 21.8 95.7 21 356.8 79 263.7 59 18,832 48 3.09 (0.31,11.45)EM 365.9 17.8 47.2 13 319.6 87 287.2 79 9,446 24 2.02 (0.14,8.39)M 487.9 23.7 32.4 7 456.0 93 402.7 83 7,605 19 0.54 (0.04,4.0)ML 288.5 14.0 11.7 4 276.9 96 252.5 88 2,945 7 0.10 (0.02,1.17)L 413.3 20.1 10.0 2 403.4 98 331.9 80 2,564 6 0.05 (0.01,0.71)Analyzed 2014.9 97.8 197.0 10a 1812.8 88a 1538.0 75a 41,392AGPv3 total 2059.9 39,475aPercentage of whole-genome coverage.bPercentage of RT class coverage.cGenes that overlap a RT class segment boundary are counted in both classes.dPercentage of FGS genes.eMedian calculated by excluding genes with FPKM = 0; values in parentheses are first and third quartiles.

DNA Replication Timing Program in Maize 2133

20 families, which display replication timing trends representativeof the rest in their group, are in Figure 5D, while results for all20 families are shown inSupplemental Figures8B to8D.Again,weusedpermutation analysis to assess if the enrichment of elementsfrom each of the top six families found in each RT class wasstatistically significant (permutation P value = 0.001) comparedwith that expected by chance. We found that the two highestabundance RLC/copia families, Ji and Opie, are significantlyenriched in the E, EM, andM classes (percentage of overlaps andasterisks indicated in Figure 5D). The four most abundant RLG/gypsy families displayed different trends. The RLG Huck family issignificantly enriched in theEandMclasses, but not the rest of theRT classes. This enrichment in noncontiguous RT classes mayrelate to the presence of several Huck subfamilies with differentcharacteristics (SanMiguel andVitte, 2009). TheRLGXilon-diguusfamily is significantly enriched in the M, ML, and L classes, andfinally the RLG Cinful-zeon and Flip families are significantly en-riched in theML and L classes (asterisks in Figure 5D). All of theserelationships are also significant in the reverse permutation tests(shuffling LTR-retro family members; see Methods). However, inthese reverse tests, the M class is also significantly enriched inRLG Cinful-zeon elements, even though RLG Cinful-zeon ele-ments are not significantly enriched in M segments (compareSupplemental Tables 2 and 3).

We further investigated these top six most abundant LTR-retrofamilies to determinewhether they differ in their proximity to genes.We found that the RLC Ji and Opie families are typically closer togenes than the RLG families (Figure 5C), which was previouslyreported as a general feature of the RLC/copia superfamily(Schnable et al., 2009). For each family,we also asked if distance tothe nearest gene variedwith RT class. Interestingly, the distributionof distance to the nearest gene for elements within a particular RTclassdoes not varywidely between families, and elements in earlierreplicating classes are consistently closer to genes (Figure 5E). Forexample, for elements found inE segments, themediandistance tothe nearest gene ranged from8 to13kb indifferent families,while inL segments the median ranges from 49 to 68 kb (Figure 5E). Al-thoughthedistancedistributionsaresimilar fordifferent families, it isimportant to note that the number of elements in each RT classvaries widely between the families (see the numbers above theboxes in Figure 5E). This result further emphasizes the apparentinsertion and/or retention biases of these families.

Maize Functional Centromeres Replicate Predominantlyin Mid S

Maize centromeres also contain highly repetitiveDNAsequences,including several centromeric retrotransposons (CRM1 andCRM2)

Figure 4. Association of Genes and Gene Transcription with Replication Time.

(A)Thenumber of 1-kbwindows in eachRTclass that overlaps anFGSgene, displayed aspercentageof all 1-kbwindows in that class. Asterisks denoteRTclasses inwhich the indicatedpercent overlapwassignificantly greater thanexpectedbychance (permutationPvalue=0.001; seeMethods). For full detailsof the permutation analysis, see Supplemental Tables 2 and 3.(B)Thedistance fromagiven gene to the nearest neighboring genewasmeasured from the59 and 39 endsof genes found in eachRTclass. This analysis didnot consider the RT class of the neighboring genes. The distribution of gene distances is shown as a box plot, with whiskers representing 1.53 IQR.(C)Gene expression values in FPKMwere calculated fromRNA-seq data for genes in each RT class, and the distribution is shown as a box plot, excludinggenes with an FPKM of zero.(D) and (E) Genes were further categorized into the following expression levels: FPKM = 0; >0 and #1; >1 and #10; >10 and #100; and >100.(D) The number of genes found in each RT class and expression group is shown.(E) The gene count from (D) is presented as the percentage of the total number of genes in each expression level group (totals shown at top of graph). SeeFigure 2I for RT class color legend.

2134 The Plant Cell

and theCentC tandem repeat (Zhong et al., 2002;Wolfgruber et al.,2009). However, functional centromeres can only be clearly dif-ferentiated frompericentromeresbybinding of centromeric histoneH3 (CENH3), a histone H3 variant (Gent et al., 2012). Using thereported locations of CENH3 binding domains (Gent et al., 2015;Zhao et al., 2016), we found that inmost cases the centromere corereplicates predominantly in the M class and is flanked by MLsegments that transition intoLsegmentsat theedgesof theCENH3binding regions (Figure 6A). Interestingly, close inspection un-covered low levels of early replication activity in the functionalcentromere (Figure 6A, blue early S replication intensity track),resulting in some small EM segments in some of the centromeres.

The predominantly M and ML replication pattern is clearest in thefully assembled centromeres of chromosomes two and five (Figure6A; Supplemental Figure 9), but holds true with some variationacross all centromeres except for chromosome one, which isparticularly poorly assembled with an extremely small mappedCENH3 binding domain of ;10 kb (Gent et al., 2015).

Analysis of Tandem Repeat Sequences in Repli-Seq Reads

Like many other large genomes, the maize genome containsa large number of tandemly repeated arrays (Plohl et al., 2008).Some of the high-copy tandem repeats in maize include the

Figure 5. Replication Times for TEs.

(A) The percentage of total TE coveragewas calculated for RT class segments (seeMethods). The distribution of percentage of coverage values in each RTclass is shown as a box plot (see Methods).(B) The total coverage inmegabases of the top 20most abundant LTR-retro families in eachRT class. The percentage of eachRT class that is composed ofthese top 20 families is shown inside the bars.(C) The distance from individual LTR-retro family members to the nearest neighboring gene was measured for the top six most abundant families, and thedistribution is shown as a box plot.(D)The coverage inmegabases of individual families from the top sixmost abundant LTR-retro families in eachRT class (xaxis sharedwith [E]). The familiesare groupedbased on their RT class abundance (“earlier,” “middle,” and “later”) and then ordered by total abundance. Asterisks denoteRT classes inwhichthe observed percentage of overlapwith each family, as indicated inside the bars, was significantly greater than expected by chance (permutation P value =0.001; seeMethods). For full details of the permutation analysis, see Supplemental Tables 2 and 3. The RT class coverage inmegabases of the top 20mostabundant LTR-retro families is shown in Supplemental Figures 8B to 8D.(E)Thedistributionofdistance to thenearest gene for familymemberswithineachRTclass from (D). Thenumberof familymembers found ineach family andRT class is indicated above the boxes. Box plot whiskers represent 1.53 IQR. See Figure 2I for RT class color legend.

DNA Replication Timing Program in Maize 2135

knob-associated knob180 and TR-1 repeats (Peacock et al.,1981; Ananiev et al., 1998a), the 5S and 45S rDNA repeats (Rivinet al., 1986), and the centromere-associated CentC repeat(Ananievet al., 1998b).Despite the largeamountofDNAestimatedto physically be in these tandem repeat arrays (30-Mb knob180,2-MbTR-1, 35-Mb45S rDNA, 3-MbCentC [Schnable et al., 2009],and ;0.8-Mb 5S rDNA [Rivin et al., 1986]), they are not wellrepresented in the genome assembly (Schnable et al., 2009).Regions that match these repeat sequences are often visible in ourraw mapped Repli-seq reads as large spikes of “collapsed” signal(e.g., seeFigure1F, arrowheads).Given that readsmapping tosuchregions are derived frommany identical copies that are sometimesfrom different locations, we excluded them from our replicationintensity profiles and segment classification. To investigate thereplication times of these biologically important tandem repeatarrays,weusedadifferentapproach that took intoaccount all of theRepli-seq sequencing reads, even those that did not uniquely mapto the genome assembly. We used consensus sequences for theknob180, TR-1, 5S and 45S rDNA, and CentC tandem repeats(courtesy of J. Gent, K. Dawe, T. Wolfgruber, and G. Presting) toindividually query trimmed Repli-seq reads using BLAST softwareand a nonstringentE value to allow for variants of each repeat (Gentet al., 2014). The resulting read counts for each tandem repeat typein each S phase or the G1 genomic DNA sample were then nor-malized to the total number of reads in that sample to obtain thepercentage of reads that align to each tandem repeat.

As expected, we found that reads corresponding to 45S rDNAand knob180 repeats are much more abundant than thosecorresponding to 5S rDNA, TR-1, and CentC (Figure 6B). Toaddress this difference, the percentage of early, mid, and latereads matching each tandem repeat was normalized to thepercentage in G1 genomic DNA to obtain a measure of the foldenrichment relative to G1 (Figure 6C). The 5S rDNA repeat hasover 3-fold enrichment in early S readsand is depleted inmid andlate S reads. The opposite pattern was clear for the 45S rDNArepeat sequence, which showed an almost 2-fold enrichmentover G1 in the late S reads. A much smaller fraction of the 45SrDNA replicates in early and mid S and likely corresponds toa small fraction of 45S copies that are transcriptionally active(Buescher et al., 1984; Tucker et al., 2010). Both of the knobrepeat sequences, knob180 and TR-1, replicate almost entirelyin the late S fraction, where they are enriched by 1.7- and 3-fold,respectively, relative to G1. This result is consistent with ourprevious cytological analysis of late S nuclei using a fluores-cence in situ hybridization probe for knob180 repeat clusters(Basset al., 2015), andwithanolder report that knobsaresomeofthe last sequences to replicate in maize (Pryor et al., 1980). Fi-nally, the CentC repeat sequence is much more distributedacross the reads in all three S phase fractions. Although most ofthis sequence replicates with the late S fraction, significantportions replicate in mid S and even early S fractions. Althoughthe distribution of the CentC repeat in our Repli-seq reads was

Figure 6. Replication Times for Centromeres and Tandem Repeat Sequences.

(A) The functional centromere of chromosome 5, as defined by CENH3 binding (black rectangle; from Zhao et al., 2016), replicates predominantly inM andtransitions to ML and L near the ends of the CENH3 binding region. See Figure 2I for RT class color legend.(B) and (C) Tandem repeat consensus sequenceswere blasted against the trimmed Repli-seq reads, independent of mapping to the reference genome, toestimate the abundance of these tandem repeat sequences in the Repli-seq reads (see Methods).(B) The percentage of reads corresponding to each tandem repeat sequence in each replication time sample.(C)The fold enrichment of each tandem repeat relative to the amount inG1. Themeanand SD (error bars) for threebiological replicates of early andmidSandtwo biological replicates of late S are displayed.

2136 The Plant Cell

initially surprising, it may reflect the presence of some smallerclustersofCentC repeatsoutsideof centromeric regions (Bilinskiet al., 2015) and our observation that the functional centromerereplicates primarily in the M and ML RT classes with loweramounts of activity in E and L.

Association of Other Chromatin Features withReplication Time

To explore potential associations of chromatin structure withreplication time, we used chromatin immunoprecipitation se-quencing (ChIP-seq) to profile the genomic locations of threehistone modifications in G1 nuclei. The histone marks wereacetylation of lysine 56 (H3K56ac), trimethylation of lysine4 (H3K4me3), and trimethylation of lysine 27 (H3K27me3).H3K4me3 is a euchromatic mark associated almost exclusivelywith genes and is highly conserved among eukaryotes (Fuchsand Schubert, 2012). H3K4me3 also shows a consistent dis-tribution in cytologically visible euchromatic regions in cells fromvarious zones of the developing maize root (Yan et al., 2014).H3K56ac has roles in several different biological processes,including transcription,DNA replicationand repair, andnucleosomedynamics in yeast and other eukaryotes (Masumoto et al., 2005; Xuet al., 2005;Hanet al., 2007;Kaplanet al., 2008;Williamset al., 2008).H3K27me3 is a mark for facultative heterochromatin and is asso-ciated with repressed transcription, developmental gene regulation,and imprinting in maize (Wang et al., 2009; Makarevitch et al., 2013;Zhang et al., 2014). We expected H3K4me3 and H3K56ac to markopen chromatin in the euchromatic portion of the genome andH3K27me3tomarkchromatin that ismorecondensed, thoughnotastightly packaged as constitutive heterochromatin. We obtained over41 million mapped read pairs for each histone mark (SupplementalTable 1). Globally, the distribution of called peaks for H3K4me3 andH3K56ac closely follows the gene distribution on maize chromo-somes (Supplemental Figure 10A). The distribution of H3K27me3also parallels the gene distribution to some degree, in that it is most

abundant near the ends of chromosomes, but it is also abundant allacross the chromosome arms except for near the centromere(Supplemental Figure 10A). Supplemental Table 4 is a summaryof the number and average size of called peaks in G1 cells fromroot tips.We then identified the replication time of regions of the

genome containing each of the three histone marks or anycombination in close proximity (within 1 kb), giving each 1-kbwindow a histone mark “signature.” We found that H3K56acand H3K4me3 by themselves or in combination have a strongassociation with earlier replicating regions, and H3K4me3 israrely found without H3K56ac in close proximity (Figure 7A).Permutation analysis also showed significant enrichment(permutation P value = 0.001) for these marks in E and EMsegments compared with that expected by chance (asterisksin Figure 7A; Supplemental Table 2). Altogether, 1-kb windowsmarked with H3K56ac and H3K4me3 peaks alone or togethercomprise 13%of the regions classified as E segments, with themajority (71%) overlapping within 1 kb of a gene. Regionsmarked by H3K27me3 are less abundant (;4–7% of each RTclass) and follow two different replication time trends. WhenH3K27me3 is colocated with one or both of the active marks itis significantly enriched in theE andEMsegments. Conversely,when H3K27me3 is alone, there is a small but significant en-richment in the ML class (asterisks in Figure 7A). However, thereverse tests, assessing the enrichment of RT classes inH3K27me3 peak regions, sometimes yielded different sig-nificance outcomes (compared in Supplemental Tables 2 and3), further indicating the complexity of the relationships withthis mark.Wecalculated themedianFPKMvalues forFGSgene-containing

windows with each histone mark signature to test whether theexpectedexpressionpatternsarepresent.Asexpected, geneswithH3K56ac or H3K4me3 alone or in combination have relatively highmedian expression levels (Supplemental Figure 10B). However,within the group of windows containing these active marks, gene

Figure 7. Replication Times for Chromatin-Related Features.

(A) The number of 1-kbwindows in each RT class that overlaps called peak regions for H3K56ac, H3K4me3, H3K27me3, or any combination of these threemarks, presented asapercentageof the total number of 1-kbwindows in eachRTclass. The inset shows thepercentageof a lower abundancehistonemarksignature, H3K4me3 without H3K56ac or H3K27me3, on an expanded y axis. The histone mark signature labeled H3K56ac/H3K4me3/H3K27me3represents any combination of either H3K56ac or H3K4me3 with H3K27me3.(B)MNase hypersensitivity (HS) region data fromwhole shoots and roots fromRodgers-Melnick et al. (2016) were overlaid with the segmented RT classesand the number ofHS regions counted in eachRTclass. The count ofHS regions permegabase coveredby eachRTclass is displayed. Asterisks denoteRTclasses inwhich theobservedpercentageofoverlapof the indicated featurewassignificantlygreater thanexpectedbychance (permutationPvalue=0.001;see Methods). For full details of the permutation analysis, see Supplemental Tables 2 and 3. See Figure 2I for RT class color legend.

DNA Replication Timing Program in Maize 2137

expression is also associated by the replication time of the genes,such that genes inE segments havehighermedian expression thanthose inM segments. Conversely, geneswith H3K27me3 aloneor in combination with either of the two active marks have zeroor extremely low median expression levels regardless of RT(Supplemental Figure 10B).

To further investigate the relationship between RT and chro-matinaccessibility,wealsocompared theRTclasseswith recentlypublished data for micrococcal nuclease hypersensitivity (MNaseHS) regions in whole shoots and roots of 9-d-old maize B73seedlings (Rodgers-Melnick et al., 2016). In maize, MNase HSregions are indicators of open chromatin that are enriched inthe areas surrounding genes, are associated with meiotic recom-bination hot spots, and explain a large amount of heritable phe-notypic variation (Vera et al., 2014; Rodgers-Melnick et al., 2016).We counted the number ofMNaseHS regions in genomic regionscorresponding to our RT class segments (Figure 7B) and alsocalculated the percentage of overlap of each class with HS re-gions. We found that MNase HS regions reported for both thewhole shoot and root samples are significantly enriched in E andEM segments (permutation P value = 0.001; Supplemental Table2) compared with values expected by chance. There was a sharpdecrease in HS regions in M and later segments (Figure 7B).Additionally, we found that 54% of HS regions are located in Esegments, and another 23% are located in EM segments(Supplemental Table 3), further supporting their significant as-sociation with early replication.

DISCUSSION

Characterizing DNA Replication in Maize

We used a novel adaptation of the Repli-seq assay by labelingnewly replicated DNA with a short EdU pulse and sorting nucleibased on EdU incorporation (as AF-488 fluorescence) in additionto the traditional DNA content. The mild conditions and rapiddetection of EdU by click chemistry (Salic and Mitchison, 2008;Darzynkiewicz et al., 2011) are an improvement to the classicalBrdU labeling protocol we used previously (Lee et al., 2010). Theuse of EdU also enables another level of purification of S phasenuclei by two-color sorting before immunoprecipitation of labeledDNA. This additional purification reduces unlabeled DNA con-tamination of the immunoprecipitates, especially in the case ofearly and late S samples, which otherwise would contain a largeexcess of unlabeled nuclei.In the course of our work, we also developed a novel and

publically available computational pipeline called Repliscan foranalyzingRepli-seqdata. This analysis pipeline automatically setsparameters based on the data themselves, is specifically tailoredto identify regions showing heterogeneous replication times, andcan be applied to data from different species (Zynda et al., 2017).With these tools, we characterized the whole-genome replicationtiming program for maize, an important crop species with a se-quenced genome two-thirds the size of the human genome(Lander et al., 2001; Schnable et al., 2009). Our use of root tips

Figure 8. Models of DNA Replication Timing Progression in Maize.

(A)Replication timing intensity profiles for early (blue), mid (green), and late (red) S-phase cells, as described in Figures 1 and 2, are overlaid to highlight thespreading pattern over consecutive fractions of S phase. Two representative regions from chromosome 5 are shown, one in themiddle of the chromosomearm (left panel) and a second in the pericentromere (right panel). Tracks containing annotated regions for total TEs, the top six most abundant LTR-retrofamilies from Figure 5 (LTR-retro), and genes, as well as a segmentation track showing the predominant replication time (RT class) are also included forreference.(B) and (C) Two nonmutually exclusive models for how replication proceeds through S phase in maize. In both models, replication begins at origins orinitiation zones (circles) and proceeds bidirectionally (arrows). In the “cascade”model (B), replication initiates in early S and cascades to adjacent originsinitiating inmid and then late S phase. In the “elongation”model (C), replication initiates at origins in early S and proceeds throughmid S regions by passiveelongation of replication forks. In thismodel, there are no origins initiating specifically inmidS phase. In the pericentromere, which predominantly replicatesin late S, the elongation model envisions that small regions with early initiation could passively elongate through mid S, followed by a second round ofinitiation events in late S phase.

2138 The Plant Cell

allowed us to analyze the replication program in a naturally oc-curring, intact organ with none of the manipulations involved increating a cell culture system. Although we cannot separate thedifferent cell types in the terminal 1mmof the root at this time, ouranalysis pipeline identified the predominant replication time forany given locus and thus provided aconsensus viewof replicationprograms in actively dividing cells of the root tip. Both the rawsequencing data and processed replication timing data files areavailable, and the processed files can be visualized using the linkto download a preformatted IGV session file (see Methods andSupplemental Tables 5 and 6).

Temporal Order of DNA Replication

Sorting three fractions of the mitotic S phase and displaying thedata from each fraction separately uncovered a complex repli-cation timing program. This program includes regions of distinctearly, mid, and late S phase replication timing, as well as manyregions that exhibit significant replication activity inmore thanoneportion of S phase. Viewed globally, the highest intensity of earlyreplication is at the gene-dense ends of chromosome arms, thehighest intensity of late replication is in the pericentromericregions, and thehighest intensity ofmid replication is locatedmid-way between the other two times. However, at the local level, weobserved many fine scale patterns of interspersed early, mid, andlate replication. In many cases, the patterns of interspersion in-dicated that sequences are replicating over consecutive fractionsof S phase in a continuous, spreading pattern, as was alsoreported for some mammalian studies that sorted more than twofractions of S phase (Chen et al., 2010; Hansen et al., 2010). Inother regions, such spreading is less apparent and, instead, thereis significant replication at more than one time in S phase. Thistiming heterogeneitywould not havebeencaptured if thedata hadbeen expressed as a ratio of early to late replication like manyearlier metazoan studies (Hiratani et al., 2008, 2010; Schwaigeret al., 2009; Ryba et al., 2010).

There are several possible sources for this heterogeneity, eachrequiring its ownspecificexperimental approach to detect. First, itis well documented from single DNA fiber studies that individualorigin use and firing efficiency can vary from cell to cell and fromone cell cycle to the next in animal systems (reviewed in Hyrien,2016). From population averaging studies of replication time, it isthought that this flexibility in origin use is constrained to a broadlydefined time in S phase (e.g., early versus late) by a higher level ofregulation in large regions termed replicationdomains (reviewed inKleinandGilbert, 2016).However,whenmore than two fractionsofS phase have been analyzed in human or mouse cell line pop-ulations, it has been estimated that up to 10% of the genomereplicates at more than one time (Farkash-Amar et al., 2008). Asecondpossible sourceof heterogeneity arises from thepresenceof multiple cell types in the root meristem region, as individual celltypesmight have different timing programs. Between any two celltypes of human or mouse differentiated embryonic stem cells, upto 20% of the genome changes replication time (Hiratani et al.,2008; Gilbert et al., 2010; Hansen et al., 2010; Ryba et al., 2010). Ithas also been reported that 12%of the genome of human primaryerythroblasts exhibits allele-specific differences in replicationtime, highlighting a third potential source of timing heterogeneity.

Such regions can be up to ;4 Mb in size and are enriched inimprinted genes (Mukhopadhyay et al., 2014). In our root tipsystem, we classified ;32% of the genome as replicating atmore than one time during S phase. This level of heterogeneitysuggests that cellular and/or allelic heterogeneity are importantin the maize root tip system. We cannot yet distinguish betweenthese two possibilities. In the future, analysis of separated celltypes, as well as comparisons with other meristems in the maizeplant, may better resolve the complexity of the replication timingprogram.

Regions of Coordinate Replication

Large regions of mammalian genomes containing multiple repli-cons have been reported to exhibit coordinate replication. Theseregions (termed replication domains) average 400 to 800 kb butcan be up to several megabases in size (Hiratani et al., 2008; Rybaet al., 2010; Klein and Gilbert, 2016). In our replication timingprofiles, we can find a few examples of coordinate regions on thisscale (e.g., the large central peak in Figure 2H), but regions of thissize are not typical for maize. Instead, regions of coordinatereplication inmaizeareusuallymuchsmaller (;50–300kb;Figures2G and 8A) and similar in size to regions reported for Drosophilaand Arabidopsis (MacAlpine et al., 2004; Schwaiger et al., 2009;Lee et al., 2010). The smaller (;30–450 kb) coordinate regions inthese species were attributed originally to their small genomesizes (Lee et al., 2010; Rhind and Gilbert, 2013). However, themaize genome is comparable in size to mammalian genomes andalmost 20 times larger than those of Drosophila and Arabidopsis,suggesting that differences in coordinate region size cannot besolely a function of genome size.It is difficult to know to what extent estimates of the size of

coordinate regions are related to differences in the scale or res-olution of the analysis, as opposed to true biological differences ingenomearrangement or regulation.Our datawerehandled in 1-kbstatic windows across the genome and only lightly smoothed,providing an in-depth, highly granular view of the complexity ofreplication timing. A similar view of mammalian replication mightreveal more complex patterns within replication domains. How-ever, there are substantial differences in the structure of the genesandgenomes inmaize versus humansormice thatmaycontributeto the observed differences in replication patterns. For example,maize differs from mammals in total gene density on chromo-somes, gene length, and intron length (Lander et al., 2001;Rafalskiand Morgante, 2004; Schnable et al., 2009). In addition, maizegenes are concentrated near the ends of chromosome arms(Schnable et al., 2009; Wang et al., 2016; Figure 2A), while humangenes are clustered in regions of increased gene density andexpression (“ridges”) dispersed across chromosomes (Caronet al., 2001; Versteeg et al., 2003). These differences in chro-mosome organization should be considered together with otherlarger scale differences in the spatial organization of replicationwithin the nucleus. In maize, mid S replicating loci do not clusteraround the nuclear and nucleolar periphery (Bass et al., 2015) astheydo in themammalian nucleus (Dimitrova andBerezney, 2002;Panning andGilbert, 2005; Zink, 2006). Instead, numerous foci aredistributed throughout the nucleoplasmduring both early andmidS phase (Bass et al., 2015).

DNA Replication Timing Program in Maize 2139

Because mammalian replication domains sometimes exhibitcoordinated changes in replication time during development(Gilbert et al., 2010), it may eventually be possible to identifysimilarly coordinated domains in maize by comparing replicationtiming programs in other maize meristems, tissues, or cell types.Such ananalysiswould provide insight into thepossible existenceand importance of large replication domains in maize.

Replication Timing in Relation to Chromatin Packaging

In eukaryotes, there is no single chromatin feature that correlatesexactly with replication time, but instead complex interactionsbetween the ensemble of histone modifications and chromatinbinding proteins are thought to relate to replication timing (RhindandGilbert, 2013).However, in all higher eukaryotes studied, earlyreplication is usually associated with active chromatin mod-ifications (Karnani et al., 2007; Schwaiger et al., 2009; Lee et al.,2010;Rybaetal., 2010;Eatonetal., 2011;Mechali et al., 2013).Wefound a similar association of early replication and active mod-ifications in maize. Many active modifications (e.g., H3K4me2/3,H3K36me3, and H3ac) are typically found in the same regions ofthe genome (Roudier et al., 2011; Julienne et al., 2013) and areusually more indicative of gene expression than replication time(Ryba et al., 2010). By selecting H3K4me3 andH3K56ac, we havelikely captured a reasonable consensus of the open regionscontaining active histone marks within the cell types in the maizeroot tip. However, it is important to note that only a small fraction(16%) of E class regions are marked by H3K4me3 or H3K56ac,suggesting that the presence of these active marks is not nec-essary for early replication to occur.

Chromatin accessibility, indicated by susceptibility to endo-nuclease cleavage, is also associated with early replication inmetazoans (Gilbert et al., 2004; Audit et al., 2009; Hansen et al.,2010; Takebayashi et al., 2012; Rhind andGilbert, 2013). Inmaize,we saw a strong association between early replication andpublished data for MNase HS sites profiled from either root orshoot tissues of 9-d-old maize B73 seedlings (Rodgers-Melnicket al., 2016). This result seems highly significant in light of the factthat in human cells, a simple model of DNA replication producedextremely accurate predictions of replication timing profiles whenDNase IHSsiteswere used to construct a “probability landscape”for initiation (Gindin et al., 2014). Additionally, other studies havenoted thatDNase I orMNaseHSsites are enriched in or near someclasses of origins in humans and yeast (Audit et al., 2009;Rodriguez and Tsukiyama, 2013; Mukhopadhyay et al., 2014;Cayrou et al., 2015).

There is less consensus on the association of late replica-tionwith repressive chromatinmodifications (e.g., H3K27me3,H3K9me2/3, andH3K20me3) (e.g., Lee et al., 2010; Julienne et al.,2013; Rhind and Gilbert, 2013) and observed associations varyamongcell types (Hiratani et al., 2008;Rybaet al., 2010;RhindandGilbert, 2013). For example, the repressive mark H3K27me3 hasbeen reported to associate with early to mid replication and withearly and mid origins (Chandra et al., 2012; Julienne et al., 2013;Picard et al., 2014), as well as with late replication (Thurman et al.,2007). In our data, peaks of H3K27me3 appear to follow twodifferentpatterns.WhenH3K27me3 is incloseproximity to the twoactivemarks, it is enriched in theEandEMclass.Alternately,when

H3K27me3 is not colocated with active marks, it is more evenlydistributed across all RT classes. Interestingly, genes containingboth of these histonemark signatures (H3K27me3with or withoutactive marks) have repressed gene expression (SupplementalFigure 10). These observations are consistent with the replicationof facultative heterochromatin during any portion of the S phase,not just in late Swhen constitutive heterochromatin has long beenknown to replicate (Pryor et al., 1980).

Diversity of Replication Timing in TE Families

TEs, which typically contain high levels of DNA methylation andthe repressive histone modification H3K9me2 in maize (Eichtenetal., 2012;Regulski etal., 2013;Westet al., 2014), are traditionallythought of as silenced chromatin. Specifically, all of the six mostabundant LTR-retro families that we investigated exhibit highoverall levels of internal DNA methylation and H3K9me2 in B73aerial tissues, as well as spreading of these heterochromaticmodifications into adjacent regions of ;1 to 2 kb (Eichten et al.,2012). Nonetheless, members of many of the highly abundantLTR-retro families are closely interspersed with genes in the eu-chromatic arms of maize chromosomes (SanMiguel et al., 1996;Liu et al., 2007; Baucom et al., 2009). Interestingly, in our data wefound that twoof thesehighlyabundant LTR-retro families,RLCJi,andOpie, were significantly enriched in E, EM, andM regions, anda third family, RLG Huck, was significantly enriched in E and Mregions. This observation appears contrary to the idea that earlyreplicating regions associate with genes, open chromatin, andactivehistonemarks.However, themediandistance to thenearestgene for the early replicating members within each LTR-retrofamily is only ;10 kb, which is less than the size estimates ofa single replicon in monocots (34–60 kb; Van’t Hof, 1996). Thisresult suggests that the earlier replicating elements in each familyrepresent a subset of gene-proximal elements that replicate inassociation with their neighboring genic regions. Whether or notthese earlier replicating subsets still maintain the same levels ofheterochromatic modifications as the rest of the family is an in-teresting question that will require further investigation. However,in Arabidopsis and rice (Oryza sativa), DNA methylation levels areknown to vary among individual LTR-retro elements as a functionof family, genomic location, age, and length, and these patternscanbemaskedwhenaveragingmethylationacrossmanydifferentinsertions (Hollister and Gaut, 2009; Vonholdt et al., 2012). Thesestudies suggest that similar variation in the level of heterochro-matic modifications may also exist in maize LTR-retro families,which could contribute to some of the differences in replicationtime. In future investigations, replication timing data may bea useful tool to help sort out different functional classes or modesof TE silencing regulation, even within a given family.

Models for Maize DNA Replication Timing

When comparing consecutive fractions of S phase, many regionsof the maize genome show a clear pattern of early replicationspreading bidirectionally into neighboring parts of the genomeduringmidSphase. In these regions, the progress of replication inmid S can be envisioned as elongation from origins that initiate inearly S and/or as initiation and elongation specific tomid S phase.

2140 The Plant Cell

Figure 8 shows a model in which these two possibilities arediagramed in two example regions, the middle of a chromosomearm, and a pericentromere. While not all initiation regions areassociated with peaks in a population average replication timingprofile, sharp peaks reflect a population preference for initiation inthat region (Yanget al., 2010;Hyrien, 2016). Thepresenceof somerelatively sharp peaks in mid S suggests these may be initiationzones (Figure 8B). However, mid peaks are generally lower in-tensity with gentler slopes than the peaks found in early S, con-sistent with the idea that high efficiency origins fire in early Sfollowed by a “cascade” of lower efficiency origins in mid S(Guilbaud et al., 2011). The generally lower intensity mid S patternis also consistent with the possibility that many mid S loci arepassively replicated by elongation from earlier initiation events(Figure 8C). In this secondmodel, there is also apotential for largerregions to be replicated by unidirectional forks, as has also beenhypothesized for TTRsbetweenearly and late replicating domainsof mammals (Farkash-Amar et al., 2008; Hiratani et al., 2008;Desprat et al., 2009). Our data do not distinguish between the twomodels, and it is possible that both scenarios occur in differentregionsof themaizegenome. If theelongationmodelwere toapplyacross large portions of the genome, we would expect to findsomeplaceswhere single replicons aremuch larger than those sofar reported in the plant literature (Van’t Hof, 1996). Detecting suchlarge replicons is technically challenging (Berezney et al., 2000).However, as single-cell sequencing and related technologiesimprove, we may be able to address this question more directly.

In a previous cytological analysis, we observed that maizeeuchromatin exists as an intermingled mixture of componentsreplicating in early and mid S phase, with the mid S componentsexhibiting a higher condensation state. We hypothesized that thispattern might reflect an alternation along the chromosome ofgene-rich regions with an extended chromatin structure andmostly replicating during early S, followed by intergenic repetitiveregions replicating during middle S phase (Bass et al., 2015). Ourmolecular data are generally consistent with this model. Earlyreplicating regionscontain thehighest concentrationof genesandhighly expressed genes (with some gene-proximal TE familymembers), while mid replicating regions are relatively more genepoor and TE dense. Genes in mid replicating regions also showgenerally lower expression. Early replicating regions are alsoenriched for histone marks associated with active chromatin(H3K4me3 and H3K56ac) and MNase hypersensitive sites, in-dicative of amoreopenchromatin structure. Finally, peaksof earlyand mid replication are arranged in an alternating pattern acrosslarge portions of the chromosome arms (Figure 8A, left panel).However, our genomic data also suggest the presence of regionswith intermediate replication time and chromatin structure, whichmay reflect gradients of chromatin accessibility between thecytologically detectable compartments.

In summary, Repli-seq data are a new class of genomic dataavailable to the maize research community. Replication timing isassociated with multiple different genomic and epigenetic fea-tures, including gene expression, chromatin accessibility, and thespatial organization of chromatin in the nucleus. In metazoans,replication timing is considered a functional readout of the manyfactors affecting large-scale chromatin structure (Rivera-Muliaet al., 2015; Rivera-Mulia and Gilbert, 2016). Recalling that the

spatial arrangement of replication activities in maize nuclei is quitedifferent from human cells (Bass et al., 2015; Savadel and Bass,2017), it will be of particular interest to examine the extent to whichmaize replication timing is regulatedby large-scale featuressuchaschromatin domains, as opposed to local features such as hyper-sensitive sites. Such questions will best be approached througha comparative analysis of maize cell types, developmental states,and genetic variants. With the rapidly growing resources availablefor othermaize lines (Luetal., 2015; Andorf et al., 2016;Hirschet al.,2016; Jiao et al., 2017), the ability to integrate the multiple kinds ofinformation represented in replication timing profiles will greatlyfacilitate comparative analyses of the maize pan-genome.

METHODS

Plant Material

Maize (Zea mays) inbred line B73 seeds were imbibed overnight, surfacesterilized, and germinated in Magenta boxes at 28°C under constant, dimlight (;500 lux, F15T8 plant and aquarium bulb) for 3 d (Wear et al., 2016).Between 450 and 550 seedlings were pooled for each of three biologicalreplicates for the Repli-seq experiments. Biological replicate material wasgrown independently and harvested on different days. After 3 d of growth,the seedling roots were immersed in sterile water containing 25 mM EdU(Life Technologies) for 20 min at room temperature with gentle agitation.After rinsing well with sterile water, the terminal 1-mm segments wereexcised from primary and seminal lateral roots. The root segments werefixed for 15min in 1% formaldehyde in 13PBS, the formaldehyde reactionquenched by adding 0.125M glycine, and the roots washed three times inPBS and snap-frozen (Wear et al., 2016).

Whole-Root Confocal Microscopy

Maize seedlings were grown, EdU labeled, and fixed as described above.Roots were harvested and embedded in 5% agarose in 13 PBS and 100-mm-thick longitudinal sections made using a Vibratome. Sections werewashed and permeabilized, and the incorporated EdU was conjugated toAF-488usingaClick-iTEdUAlexaFluor 488 imagingkit (LifeTechnologies)according to the manufacturer’s instructions. Sections were counter-stained with 0.1 mg/mL DAPI in 13 PBS and imaged on a Zeiss LSM710 confocal laser scanning microscope with 405-nm (DAPI) and 488-nm(AF-488) lasers at the Cellular and Molecular Imaging Facility at NorthCarolina State University.

Nuclei Isolation

The fixed, frozen roots described abovewere ground in cell lysis buffer (CLBfrom Wear et al., 2016) supplemented with a “Complete Mini” protease in-hibitorcocktail tablet (Roche) inasmallcommercial foodprocessor (CuisinartMini-Prep Processor, model DLC-1SS) at 4°C. The resulting homogenatewas filtered and centrifuged as previously described (Bass et al., 2015;Wearetal.,2016). Isolatednucleiwerewashed inmodifiedCLBbuffer (CLBwithoutEDTA or b-mercaptoethanol), and the incorporated EdU was conjugated toAF-488 using a Click-iT EdU Alexa Fluor 488 imaging kit (Wear et al., 2016).Finally, the nuclei were resuspended in CLB containing 2 mg/mL DAPI and40mg/mLRibonucleaseAandfiltered throughaCellTrics20-mmnylonmeshfilter (Partec) just before flow cytometry and sorting.

Flow Cytometry and Sorting

Isolated, fixed nuclei used for Repli-seq experiments were sorted andrecovered with an InFlux flow cytometer (BD Biosciences) equipped with

DNA Replication Timing Program in Maize 2141

UV (355 nm) and blue (488 nm) lasers. Events were triggered on forward-angle light scatter anddata collectedusing90° side scatter and460/50-nmand530/40-nmband-passfilters (Bassetal., 2014,2015;Wearetal., 2016).Nuclei prepared fromthe terminal 1-mmroot segmentsweresorted into13NaCl-Tris-EDTA (STE) buffer, pH 7.5, using substage gates to collectpopulations of EdU/AF-488-labeled nuclei with DNA contents in threedefined gates between 2C and 4C, corresponding to early, mid, and late Sphase (Figure 1C). For each biological replicate, 0.4 to 13 106 nuclei weresorted for each fraction of S phase, and 1 3 106 unlabeled G1 (2C DNAcontent) nuclei were sorted to use as a reference. Additionally, a smallsampleof nuclei (;50,000)were also sorted fromeachgate intoCLBbuffercontainingDAPI and reanalyzed todetermine the sort purity (SupplementalFigures 1B and 1C). Flow cytometry data were analyzed using FlowJosoftware v10.0.6 (Tree Star). Plots of side scatter versusDNAcontent (460/50 nm) were used to set analysis gates that excluded cellular debris in theflow cytometry plots (Supplemental Figure 1A).

Genomic DNA Extraction from Sorted Nuclei

Formaldehyde cross-links were reversed and DNA was solubilized byincubating the sorted nuclei in 50mMEDTA, 1%sodium lauroyl sarcosine,and 230 mg/mL proteinase K for 1 h at 42°C and then at 65°C overnight inthe dark. To inactivate the proteinase K, samples were treated with 8 mMPMSF for 40 min at room temperature prior to extraction of genomic DNAusing phenol/chloroform/isoamyl alcohol and a phase lock gel (5 Prime).Theupper aqueousphasewasmixedwith 150mg/mLGlycoBlue (Ambion),and the DNAwas precipitated in 0.3 M sodium acetate and 0.6 volumes ofcold isopropyl alcohol. The DNAwas pelleted by centrifugation at 21,130gand20°C for 30min,washedwith 70%ethanol, andcentrifuged at 21,130gand 20°C again for 15 min, dried for 5 min using a SpeedVac concentrator(Savant), and resuspended in 130 mL PCR grade water. DNA was shearedusing a S2 focused-ultrasonicator (Covaris; Sonolab Simple) with 10%duty cycle, intensity of 5, 200 cycles per burst, and a 4-min cycle length toachieve an average fragment size of ;250 bp.

EdU/AF-488-Labeled DNA Immunoprecipitation

One to 2.3 mg of sheared input DNA for each IP reaction was brought toa volume of 500 mL with ChIP dilution buffer (Gendrel et al., 2005) andprecleared by slowly mixing (8 rpm) for 1 h at 4°C with 20 mL magneticprotein G beads (Dynabeads Life Technologies) preequilibrated in ChIPdilution buffer. The beadsweremagnetically captured and the supernatantwas transferred to a clean tube. The samples were then incubated over-night at 4°C with a 1:200 dilution of anti-Alexa Fluor 488 antibody (Mo-lecular Probes; A-11094, lot 895897) followed by incubation for 2 h at 4°Cwith 25 mL of preequilibrated protein G beads. The beads were recoveredwith amagnet andwashedaspreviously describedbyGendrel et al. (2005),except an additional 5-min wash was added for each wash step. EdU-labeled, newly synthesized DNA was eluted from the beads by incubatingwith 250mL of elution buffer (1%SDS and 100mMsodium bicarbonate) at65°C for 15 min. Beads were magnetically captured and the supernatantwas transferred to a new tube. The elution was repeated once moreand both supernatants combined for a final volume of 500 mL. Im-munoprecipitated DNA, yielding from 0.3 to 0.7 ng DNA per 10,000 nuclei,was purified with a QIAquick PCR purification kit (Qiagen) following themanufacturer’s instructions and was eluted from the QIAquick columns in32 mL elution buffer. The Alexa-488 DNA IP efficiencies ranged from 1 to2.2%.

Library Construction and Sequencing for Repli-Seq Experiments

Repli-seq paired-end libraries were constructed from 5 to 10 ng of DNAusing theNEXTflex Illumina ChIP-Seq Library Prep Kit (Bioo Scientific) andthe ultra-low input protocol. After adapter ligation, the libraries were

amplified using 18 cycles of PCR. Individual samples from three biologicalreplicateswerebar-coded, pooled, andsequencedusing three lanesof theIllumina HiSeq 2000 platform.

Read Trimming and Alignment

To improve alignment rates and reduce errors, Trim Galore! v0.3.7 (http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) was used toremove 39 universal adapters from the reads, trim the 59 ends with FastQquality scoresbelow20, anddiscard reads trimmedshorter than40bp. Thequality controlled reads were then aligned to the B73 RefGen_v3 (AGPv3)genome downloaded from Ensemble Plants (Kersey et al., 2016) usingBWA-MEMv0.7.12 (Li, 2013) with default parameters. Due to the repetitivenature of the maize genome and the sensitivity of replication signals, onlyunique alignments whose orientations made them proper pairs for ourdownstream analysis were used. See Supplemental Table 1 for mappingstatistics and total sequencing coverage.

Replication Timing Data Analysis

Repli-seq data were analyzed using Repliscan as described in detail byZyndaet al. (2017). Readdensitieswere calculated in 1-kbwindowsacrossthe genome and the correlation between biological replicates was as-sessed (Supplemental Figure 2). After observing a strong Pearson cor-relation of 0.8 to 0.98 between the biological replicates of each sample, thereplicates were then summed (Zynda et al., 2017). Genomic windows withartificially high or extremely low log-transformedcoverage in the upper andlower 2.5% tails of a calculated gamma distribution were removed (Zyndaet al., 2017), and then data were normalized using sequence depth scaling(Diaz et al., 2012). In each 1-kb window, the data from each of the S phasesamples were divided by the nonreplicating G1 reference data to furthernormalize for sequencing biases. To reduce noise, without spreading peakboundaries, Haar wavelet smoothing was performed using the softwarepackage wavelets from Percival and Walden (2000). Haar wavelet levelthree was chosen because it removed low-amplitude noise, while alsopreserving replication peak boundaries.

Classifying Predominant Replication Time

The strategy and details of classifying a predominant time of replication foreach 1-kb window across the genome is described by Zynda et al. (2017).To classify a predominant time of replication, a threshold of replicationwasfirst calculated.Our experimental protocol labelsDNA that is replicating, soit is hard to discount any signal. Therefore, an automatic analysis was usedthat maximized chromosomal inclusion while also excluding low signalsfrom previously included windows. Starting from the point of the largestabsolute change in coverage (slope) for each chromosome, the replicationthreshold was lowered (increasing chromosome coverage) until the ab-solute change in coverage went below 0.1, meaning very few new chro-mosomal windowswere included if the threshold was lowered further. Thealgorithm automatically set segmentation thresholds for each chromo-some; in this case, normalized signal thresholds were between 0.84 and0.86. Only values above the threshold were considered when segmentingthe genome into predominant replication time classes. The predominanttime in which a 1-kbwindow replicates was determined by considering theproportion of total replication signal above the threshold occurring in early,mid, and late S. All signals in each 1-kb window were divided by themaximum value (infinity-norm), scaling the largest value to 1 and all othersbetween 0 and 1. A window was then classified as predominantly repli-cating in any S phase time with signal greater than 0.5. The infinity-normensured that the largest valuewas always classified as replicating, and thisclassification method allowed for a window to be called predominantlyreplicating at more than one time in S phase (e.g., both early and mid)when other signals were within 50% of the maximum value. The final

2142 The Plant Cell

classifications of predominant replication time include early (E), earlyandmid (EM), mid (M), mid and late (ML), late (L), early and late (EL), panS (EML), and not segmented (NS; unable to be called replicating at anytime).

Replication Intensity and Relative Distance from the Centromere

The normalized and smoothed replication intensity profiles for early, mid,and late S phase in Figure 2 and Supplemental Figure 6 were used tocalculate the percent of total replication in consecutive windows, eachrepresenting 10%of a given chromosomearm, andplotted as a function ofrelative distance from the centromere. Centromere positions in the B73AGPv3 genome were defined as CENH3 binding domains previously re-ported (Gent et al., 2015; Zhao et al., 2016). Centromere locations forchromosomes 1, 6, and 7 are more uncertain due to low CENH3 mapp-ability and/or poor reference genome assembly (Gent et al., 2015; Zhaoet al., 2016).

Genomic Features

For comparison with Repli-seq data, the GC content and annotations forTEs and genes were taken from the B73 genome assembly AGPv3 andaveraged across 10-kb static windows. For visual representation in IGV,these data were further smoothed using the R function ksmooth witha Gaussian (normal) kernel and a bandwidth of 5.

Gene Expression Analysis

Maize seedlings were grown as described above, with pooled root tissuefrom 45 seedlings used for each of three independently grown and har-vestedbiological replicates. The rootswere rinsedquicklywith sterilewaterand the terminal 1-mm root segments were excised, snap-frozen in liquidnitrogen, and stored at –70°C. To isolate total RNA, 10 mg of frozen rootsegments were ground in a mortar and pestle with liquid nitrogen, and thepowder was added to a tube of 0.5 mL cold PureLink Plant RNA reagent(Ambion). The manufacturer’s instructions for a small-scale RNA isolationwere followed, except that 90 mg/mL of GlycoBlue carrier (Ambion) wasaddedbefore thechloroformextractionstep. TheRNAwas resuspended in250 mL RNase-free water and stored at –70°C. Contaminating DNA wasremoved from 8 mg of total RNA using a Turbo DNA-free kit (Ambion)following themanufacturer’s instructions, except that 4mL of TurboDNasewas used and the sample was incubated with the DNase for 45 min. DNA-free RNA was collected by isopropyl alcohol precipitation, washed with75% ethanol, and resuspended in RNase-free water. DNA-free RNA wasquantified using a Qubit (Molecular Probes) and the yield was ;7 mg ofRNA. Fourmicrograms of DNA-free RNAwas processed further to depleterRNA using a Ribo-Zero Magnetic Kit for plant seed/root (EpiCentre) ac-cording to the manufacturer’s instructions. The effectiveness of the rRNAdepletion was assessed by RT-qPCR using a qScript One-Step SYBRGreenqRT-PCRkit, LowROX (QuantaBiosciences), andprimers formaize,16S, 18S, 23S, and 26S rRNA. For each biological replicate, 77 to 133 ngofrRNA-depleted RNA was used for cDNA conversion and Illumina libraryconstruction using a ScriptSeq v2 RNA-Seq Library Preparation Kit (Ep-icentre), and samples were bar-coded with ScriptSeq Index PCR Primers(Epicentre). After sequencing, Illumina adapter sequences were trimmedwith Trim Galore! (http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) and mapped to the B73 genome AGPv3 using Bowtie v2.1.0(Langmead et al., 2009) and TopHat v3.2.3 (Trapnell et al., 2009) withdefault parameters (for mapping statistics and genome coverage, seeSupplemental Table1).Mapped readsweresortedusingSAMtools v0.1.19(Li et al., 2009), and normalized gene expression values were calculatedusing Cufflinks v0.9.3 (Trapnell et al., 2010) for gene annotations from theAGPv35b+FGS.The total numberof genesoverlappingeachRTclasswasdivided into five groups based on FPKM values derived from Cufflinks:

FPKM = 0, 0 > FPKM$ 1, 1 > FPKM$ 10, 10 > FPKM$ 100, and FPKM >100 in the same manner as Regulski et al. (2013).

TE Families

Genome locations in the B73 AGPv3 annotation of LTR retrotransposonfamilies from themaize TE consortium databasewere used (Baucomet al.,2009; Schnable et al., 2009). Overlapping and redundant sequence in-tervalswerecollapsedusingacustomRscript (RDevelopmentCoreTeam,2016) (script and collapsed files courtesy of M. Stitzer). Of these LTRretrotransposon families, the top 20 most abundant families across thesuperfamilies of RLG/gypsy, RLC/copia, and RLX/unknown as defined byBaucom et al. (2009) were used for our analysis.

Tandem Repeat Sequences

Reference sequences were acquired for maize tandem repeat classes,including CentC, TR-1, knob180 (courtesy of J. Gent and K. Dawe), and 5Sand 45S rDNA (courtesy of T. Wolfgruber and G. Presting). Given that themajority of tandem repeat sequences are not included in the B73 AGPv3reference genome assembly, the abundance of these repeat sequences inourRepli-seqdata independentof the referencegenomewasmeasured,asdescribed by Gent et al. (2014), allowing us to query all reads and not justuniquely mapping ones. To do so, filtered, trimmed, and adapter-free DNAfragment reads from individual biological replicates of G1, early, mid, andlate S samples were aligned to consensus sequences for each tandemrepeat family usingBLASTsoftware (parameter “-e1e-8”). For eachsampleand biological replicate, the number of reads that aligned to each repeatfamily was normalized to the total number of reads in the sample. Finally,the relative abundance of each family in early, mid, or late reads wasnormalized to the relativeabundanceof thesamefamily in theG1reference.

ChIP-Seq Analysis of Histone Modifications

Maize seedlings were grown as described above. For each of the threeChIP-seq experiments listed below, root tissue was pooled from between430 and 720 seedlings for three independently grown and harvested bi-ological replicates. Seedling roots were pulse-labeled with 25 mM EdU for1h, and the terminal 3-mm (H3K4me3andH3K27me3) or 5-mm (H3K56ac)root segments were excised and fixed as described above. After nucleiisolation, the incorporated EdU was conjugated to AF-488, total DNA wasstained with DAPI, and then nuclei were flow sorted as described above.The unlabeled G1 (2C) nuclei were collected in 23 extraction buffer 2 (EB2)(Gendrel et al., 2005) diluted to 13with the sorted drops of 13 STE sheathfluid. The antibodies used for ChIP were anti-H3K56ac rabbit polyclonal1:200 dilution (Millipore; 07-677, lot DAM1462569), anti-H3K4me3 rabbitmonoclonal 1:300 dilution (Millipore; 07-473, lot DAM1779237), and anti-H3K27me3 rabbit polyclonal 1:300 dilution (Millipore; 07-449, lot2,275,589). ChIP procedures were adapted from Gendrel et al. (2005).Briefly, fixed, sortednuclei in EB2bufferwerecentrifugedat 12,000gat 4°Cfor 10min. Thesupernatantwasdiscarded and thepelletwas resuspendedin110mLofnuclei lysisbuffer (Gendrel etal., 2005).Chromatinwasshearedusing a S2 focused-ultrasonicator (Covaris; Sonolab Simple) with 10%duty cycle, intensity of 5, and 200 cycles per burst for 10min to achieve anaverage fragment size of ;200 bp. After shearing, the ChIP protocol ofGendrel et al. (2005) was followed from their step 17, except for the fol-lowing changes: The chromatin sample was initially brought up to 1 mL inChIP dilution buffer, Dynabeads protein G magnetic beads (Life Tech-nologies) were used, an additional 5-min wash was added for each washstep, and a treatment of 55mg/mLRNaseA at 37°C for 1 hwas added afterreversing the cross-links. The final DNA purification after the ChIP wasdone using a QIAquick PCR purification kit (Qiagen). ChIP-seq librarieswereconstructedandsequenced in thesamewayas theRepli-seq librariesdescribed above, except that 0.6 to 15 ng of DNA was used for library

DNA Replication Timing Program in Maize 2143

construction. After sequencing, Illumina adapter sequenceswere trimmedand mapped to AGPv3 as described for Repli-seq data. Redundant readswere removed from each of the BAM alignment files using PICARD (http://broadinstitute.github.io/picard/) andSAMtools (Li et al., 2009) (formappingstatistics and genome coverage, see Supplemental Table 1). EnrichedChIP binding regions, also known as “peaks,” were called using MACSv2.1.0 (Zhang et al., 2008) with parameters “–nomodel–nolambda–broad”and a q-value threshold of 0.01. The called peaks for all three histonemodifications were intersected with the RT classes using intersectBed inthe BEDTools suite (Quinlan and Hall, 2010) to assign each 1-kb windowa histone mark “signature” containing any combination of each mark. Themedian gene expression level and first and third quartile values werecomputed for 1-kb windows containing a gene and each histone markcombination in the E, M, and L RT classes using the FPKM data describedabove.

MNase Hypersensitivity Site Analysis

Published data sets of the genomic locations of MNase HS regions fromwhole root and shoot tissues from 9-d-old B73 seedlings, as described byRodgers-Melnick et al. (2016), were provided by E. Buckler. The number ofHS regions found in the genomic regions corresponding to our RT classeswere counted and normalized by the total number of megabases in eachclass. The percentage of coverage of both HS regions that overlap witheach RT class and RT classes that overlap with HS regions was alsocalculated.

Association of Genetic and Chromatin Features with ReplicationTime Classes

Various genetic and chromatin features were associated with the RTclasses to determine the overlap of a particular feature with each RT class.To calculate the coverage of genes, LTR-retro families, or histone marksignatures, the genomic locations and calculated valueswere representedin bedGraph format with the window size of 1 kb. The values for individualgenomic or chromatin features that overlapped with different RT classeswere stored using intersectBed in the BEDTools suite (Quinlan and Hall,2010). For computing the count of genes in each RT class, the GFF3 fileformat was used to identify the gene coordinates from the maize B73AGPv3 annotation and was first computed into bedGraph format and thenthe samemethodswere used as above. FPKMvalueswere also appendedonto this bedGraph file for associating expression levels with genesoverlappingwith different RT classes. Themedian gene distance for genesfound in eachRT classwas calculated bymeasuring the genomic distancefrom the59and39 endsof eachgene to thenext nearest gene, regardless oftheRTclass of the nearest gene. Thedistance fromelementswithin the topsix most abundant LTR-retro families to the nearest gene was calculatedusing closestBed (parameter –d) usingBED formattedcoordinates for bothfeatures.

Permutation Analysis

A permutation or feature randomization test, similar to that describedpreviously (De and Michor, 2011; Bartholdy et al., 2015), was used toassess the statistical significance of the observed overlap values betweenRT segment classes and other features. To test the significance of theenrichmentofvarious features (includinggenes,LTR-retro families,histonemarks, and MNase HS regions) in each of the RT segment classes, the RTsegments were randomly shuffled (Supplemental Table 2). To test thereverse relationship, namely, the significance of the enrichment of RTsegment classes in each feature, the feature was randomly shuffled(Supplemental Table 3). To do this, shuffleBed in the BEDTools suite wasused (Quinlan and Hall, 2010) with default parameters to generate1000 random genomic location lists as a null distribution for each feature or

RT class, preserving the number and size of the original intervals.Several shuffle parameterswere tested in shuffleBed to determine if thesignificance outcomes were consistent, irrespective of randomizationstrategy. These parameters included allowing or disallowing overlapsand excluding “bad spots” in the genome assembly (see replicationtiming data analysis). None of these parameters on their own produceddifferent significance outcomes; however, the combinations of severalof these parameters did produce different outcomes because theygreatly limited the shuffle step for larger features or segments. Thus, forthe final tests, the default parameters (overlaps allowed, no otherexclusions) were used, which impose minimal assumptions on the nulldistributions (De et al., 2014). The percentage overlap between the twooriginal data sets was calculated as the test statistic (observed value)using intersectBed. An empirical P value was estimated by calculatingthe proportion of N data sets (n = 1000 shuffled + 1 observed value) witha percentage of overlap value greater than or equal to the observedpercent overlap (Ernst, 2004). Permutation P values of 0.001, indicatingthat none of the randomly shuffled data sets had a percentage ofoverlap value greater than or equal to the observed value, were ac-cepted as evidence for enrichments significantly greater than expectedby random chance.

Accession Numbers

Processed data files formatted for the IGV, as well as a pregenerated IGVsession containing the data files, are available for download from theCyVerse Data Store (previously iPlant Collaborative; Merchant et al., 2016)via the links listed in Supplemental Table 5. Sequence data from this articlecanbe found in theNCBISequenceReadArchive (SRA) under the umbrellaaccession number PRJNA335625. The SRA numbers for each experimentare listed in Supplemental Table 6.

Supplemental Data

Supplemental Figure 1. Flow cytometric sorting of nuclei andassessment of purity.

Supplemental Figure 2. Correlation of Repli-seq biological replicates.

Supplemental Figure 3. Bioinformatic analysis of Repli-seq datausing Repliscan.

Supplemental Figure 4. Distribution of replication activity on chro-mosome arms.

Supplemental Figure 5. Replication activity on individual chromo-some arms.

Supplemental Figure 6. Replication intensity profiles for all ten maizechromosomes.

Supplemental Figure 7. The percent GC content distribution varieslittle between RT classes.

Supplemental Figure 8. Replication time and genomic distribution ofindividual LTR-retrotransposon families.

Supplemental Figure 9. Replication time of chromosome twocentromere.

Supplemental Figure 10. Histone mark genomic distribution andtranscriptional activity of windows with histone mark signatures.

Supplemental Table 1. Sequence mapping statistics for Repli-seqand companion data sets.

Supplemental Table 2. Percent of RT segment classes overlappingwith features and corresponding permutation P values.

Supplemental Table 3. Percent of features overlapping with RTsegment classes and corresponding permutation P values.

2144 The Plant Cell

Supplemental Table 4. Histone mark called peak region summaries.

Supplemental Table 5. Processed data availability information.

Supplemental Table 6. SRA accession numbers.

ACKNOWLEDGMENTS

Wethankpresentand former labmembers fromNCSU,RoselynHatch,AshleyBrooks, and Emily Wheeler, for assistance with material harvest and helpfuldiscussions, aswell asHankBass for helpful discussions.We thank JonathanGent, Kelly Dawe, Thomas Wolfgruber, and Gernot Presting for supplyingconsensus sequences for the tandem repeats; Michelle Stitzer for supplyingthefileswithgenomic locationsof individualLTR-retrotransposon families;andEdward Buckler for supplying the files with genomic locations of MNase HSsites. We thank Mark Millard (USDA, ARS, NCRPIS) for supplying our originalB73 seed stock (GRINNPGSPI 550473). This work was supported by a grantfrom the NSF PGRP (NSF IOS-1025830 to L.H.-B. and W.F.T.).

AUTHOR CONTRIBUTIONS

E.E.W., T.-J.L., G.C.A., R.A.M., M.W.V., L.H.-B., and W.F.T designed theresearch. E.E.W., C.L., T.-J.L., L.M.-Y., P.M., and E.S.S performed theresearch. J.S., G.J.Z., L.C., and M.W.V. contributed analytical and com-putational tools.E.E.W., J.S., andG.J.Zanalyzed thedata. E.E.W.andW.F.T.primarily wrote the article. All authors approved of the final article.

Received January 17, 2017; revised July 31, 2017; accepted August 24,2017; published August 25, 2017.

REFERENCES

Albert, P.S., Gao, Z., Danilova, T.V., and Birchler, J.A. (2010). Di-versity of chromosomal karyotypes in maize and its relatives. Cy-togenet. Genome Res. 129: 6–16.

Ananiev, E.V., Phillips, R.L., and Rines, H.W. (1998a). A knob-associated tandem repeat in maize capable of forming fold-backDNA segments: are chromosome knobs megatransposons? Proc.Natl. Acad. Sci. USA 95: 10785–10790.

Ananiev, E.V., Phillips, R.L., and Rines, H.W. (1998b). Chromosome-specific molecular organization of maize (Zea mays L.) centromericregions. Proc. Natl. Acad. Sci. USA 95: 13073–13078.

Andorf, C.M., et al. (2016). MaizeGDB update: new tools, data andinterface for the maize model organism database. Nucleic AcidsRes. 44: D1195–D1201.

Audit, B., Zaghloul, L., Vaillant, C., Chevereau, G., d’Aubenton-Carafa, Y., Thermes, C., and Arneodo, A. (2009). Open chromatinencoded in DNA sequence is the signature of ‘master’ replicationorigins in human cells. Nucleic Acids Res. 37: 6064–6075.

Baluska, F. (1990). Nuclear size, DNA content, and chromatin con-densation are different in individual tissues of the maize root apex.Protoplasma 158: 45–52.

Bartholdy, B., Mukhopadhyay, R., Lajugie, J., Aladjem, M.I., andBouhassira, E.E. (2015). Allele-specific analysis of DNA replicationorigins in mammalian cells. Nat. Commun. 6: 7051.

Bass, H.W., Wear, E.E., Lee, T.J., Hoffman, G.G., Gumber, H.K.,Allen, G.C., Thompson, W.F., and Hanley-Bowdoin, L. (2014). Amaize root tip system to study DNA replication programmes insomatic and endocycling nuclei during plant development. J. Exp.Bot. 65: 2747–2756.

Bass, H.W., Hoffman, G.G., Lee, T.J., Wear, E.E., Joseph, S.R.,Allen, G.C., Hanley-Bowdoin, L., and Thompson, W.F. (2015).Defining multiple, distinct, and shared spatiotemporal patterns ofDNA replication and endoreduplication from 3D image analysis ofdeveloping maize (Zea mays L.) root tip nuclei. Plant Mol. Biol. 89:339–351.

Baucom, R.S., Estill, J.C., Chaparro, C., Upshaw, N., Jogi, A.,Deragon, J.M., Westerman, R.P., Sanmiguel, P.J., andBennetzen, J.L. (2009). Exceptional diversity, non-random distri-bution, and rapid evolution of retroelements in the B73 maize ge-nome. PLoS Genet. 5: e1000732.

Berezney, R., Dubey, D.D., and Huberman, J.A. (2000). Heteroge-neity of eukaryotic replicons, replicon clusters, and replication foci.Chromosoma 108: 471–484.

Bilinski, P., Distor, K., Gutierrez-Lopez, J., Mendoza, G.M., Shi, J.,Dawe, R.K., and Ross-Ibarra, J. (2015). Diversity and evolutionof centromere repeats in the maize genome. Chromosoma 124:57–65.

Blow, J.J., Ge, X.Q., and Jackson, D.A. (2011). How dormant originspromote complete genome replication. Trends Biochem. Sci. 36:405–414.

Boulos, R.E., Drillon, G., Argoul, F., Arneodo, A., and Audit, B.(2015). Structural organization of human replication timing domains.FEBS Lett. 589: 2944–2957.

Bryant, J.A., and Aves, S.J. (2011). Initiation of DNA replication:functional and evolutionary aspects. Ann. Bot. (Lond.) 107: 1119–1126.

Buescher, P.J., Phillips, R.L., and Brambl, R. (1984). Ribosomal RNAcontents of maize genotypes with different ribosomal RNA genenumbers. Biochem. Genet. 22: 923–930.

Caron, H., et al. (2001). The human transcriptome map: clustering ofhighly expressed genes in chromosomal domains. Science 291:1289–1292.

Castellano, M.M., del Pozo, J.C., Ramirez-Parra, E., Brown, S., andGutierrez, C. (2001). Expression and stability of Arabidopsis CDC6are associated with endoreplication. Plant Cell 13: 2671–2686.

Cayrou, C., Ballester, B., Peiffer, I., Fenouil, R., Coulombe, P.,Andrau, J.C., van Helden, J., and Mechali, M. (2015). The chro-matin environment shapes DNA replication origin organization anddefines origin classes. Genome Res. 25: 1873–1885.

Chandra, T., et al. (2012). Independence of repressive histone marksand chromatin compaction during senescent heterochromatic layerformation. Mol. Cell 47: 203–214.

Chen, C.L., Rappailles, A., Duquenne, L, Huvet, M., Guilbaud G.,Farinelli, L., Audit, B., d’Augenton-Carafa, Y, Arnedodo, A, Hyrien,O., and Thermes, C., (2010). Impact of replication timing on non-CpGand CpG substitution rates in mammalian genomes. Genome Res. 20:447–457.

Costas, C., de la Paz Sanchez, M., Sequeira-Mendes, J., andGutierrez, C. (2011a). Progress in understanding DNA replicationcontrol. Plant Sci. 181: 203–209.

Costas, C., et al. (2011b). Genome-wide mapping of Arabidopsisthaliana origins of DNA replication and their associated epigeneticmarks. Nat. Struct. Mol. Biol. 18: 395–400.

Darzynkiewicz, Z., Traganos, F., Zhao, H., Halicka, H.D., and Li, J.(2011). Cytometry of DNA replication and RNA synthesis: historicalperspective and recent advances based on “click chemistry”. Cy-tometry A 79: 328–337.

De, S., and Michor, F. (2011). DNA replication timing and long-rangeDNA interactions predict mutational landscapes of cancer ge-nomes. Nat. Biotechnol. 29: 1103–1108.

De, S., Pedersen, B.S., and Kechris, K. (2014). The dilemma ofchoosing the ideal permutation strategy while estimating statisticalsignificance of genome-wide enrichment. Brief. Bioinform. 15: 919–928.

DNA Replication Timing Program in Maize 2145

DePamphilis, M., and Bell, S. (2010). Genome Duplication. (New York:Garland Science).

Desprat, R., Thierry-Mieg, D., Lailler, N., Lajugie, J., Schildkraut,C., Thierry-Mieg, J., and Bouhassira, E.E. (2009). Predictabledynamic program of timing of DNA replication in human cells. Ge-nome Res. 19: 2288–2299.

Diaz, A., Park, K., Lim, D.A., and Song, J.S. (2012). Normalization,bias correction, and peak calling for ChIP-seq. Stat. Appl. Genet.Mol. Biol. 11: 9.

Dimitrova, D.S., and Berezney, R. (2002). The spatio-temporal or-ganization of DNA replication sites is identical in primary, immor-talized and transformed mammalian cells. J. Cell Sci. 115: 4037–4051.

Eaton, M.L., Prinz, J.A., MacAlpine, H.K., Tretyakov, G.,Kharchenko, P.V., and MacAlpine, D.M. (2011). Chromatin sig-natures of the Drosophila replication program. Genome Res. 21:164–174.

Eichten, S.R., et al. (2012). Spreading of heterochromatin is limitedto specific families of maize retrotransposons. PLoS Genet. 8:e1003127.

Ernst, M.D. (2004). Permutation methods: a basis for exact inference.Stat. Sci. 19: 676–685.

Farkash-Amar, S., Lipson, D., Polten, A., Goren, A., Helmstetter,C., Yakhini, Z., and Simon, I. (2008). Global organization of repli-cation time zones of the mouse genome. Genome Res. 18: 1562–1570.

Fuchs, J., and Schubert, I. (2012). Chromosomal distribution andfunctional interpretation of epigenetic histone marks in plants. PlantGenet. Genomics 4: 231–253.

Galbraith, D.W., Harkins, K.R., Maddox, J.M., Ayres, N.M., Sharma,D.P., and Firoozabady, E. (1983). Rapid flow cytometric analysis ofthe cell cycle in intact plant tissues. Science 220: 1049–1051.

Gendrel, A.V., Lippman, Z., Martienssen, R., and Colot, V. (2005).Profiling histone modification patterns in plants using genomic tilingmicroarrays. Nat. Methods 2: 213–218.

Gent, J.I., Dong, Y., Jiang, J., and Dawe, R.K. (2012). Strong epi-genetic similarity between maize centromeric and pericentromericregions at the level of small RNAs, DNA methylation and H3 chro-matin modifications. Nucleic Acids Res. 40: 1550–1560.

Gent, J.I., Wang, K., Jiang, J., and Dawe, R.K. (2015). Stable pat-terns of CENH3 occupancy through maize lineages containing ge-netically similar centromeres. Genetics 200: 1105–1116.

Gent, J.I., Ellis, N.A., Guo, L., Harkess, A.E., Yao, Y., Zhang, X., andDawe, R.K. (2013). CHH islands: de novo DNA methylation in near-gene chromatin regulation in maize. Genome Res. 23: 628–637.

Gent, J.I., Madzima, T.F., Bader, R., Kent, M.R., Zhang, X., Stam,M., McGinnis, K.M., and Dawe, R.K. (2014). Accessible DNA andrelative depletion of H3K9me2 at maize loci undergoing RNA-directed DNA methylation. Plant Cell 26: 4903–4917.

Gilbert, D.M., Takebayashi, S.I., Ryba, T., Lu, J., Pope, B.D.,Wilson, K.A., and Hiratani, I. (2010). Space and time in the nu-cleus: developmental control of replication timing and chromosomearchitecture. Cold Spring Harb. Symp. Quant. Biol. 75: 143–153.

Gilbert, N., Boyle, S., Fiegler, H., Woodfine, K., Carter, N.P., andBickmore, W.A. (2004). Chromatin architecture of the human ge-nome: gene-rich domains are enriched in open chromatin fibers.Cell 118: 555–566.

Gindin, Y., Valenzuela, M.S., Aladjem, M.I., Meltzer, P.S., andBilke, S. (2014). A chromatin structure-based model accuratelypredicts DNA replication timing in human cells. Mol. Syst. Biol.10: 722.

Guilbaud, G., Rappailles, A., Baker, A., Chen, C.L., Arneodo, A.,Goldar, A., d’Aubenton-Carafa, Y., Thermes, C., Audit, B., and

Hyrien, O. (2011). Evidence for sequential and increasing activationof replication origins along replication timing gradients in the humangenome. PLOS Comput. Biol. 7: e1002322.

Han, J., Zhou, H., Li, Z., Xu, R.M., and Zhang, Z. (2007). Acetylationof lysine 56 of histone H3 catalyzed by RTT109 and regulated byASF1 is required for replisome integrity. J. Biol. Chem. 282: 28587–28596.

Hansen, R.S., Thomas, S., Sandstrom, R., Canfield, T.K., Thurman,R.E., Weaver, M., Dorschner, M.O., Gartler, S.M., andStamatoyannopoulos, J.A. (2010). Sequencing newly replicatedDNA reveals widespread plasticity in human replication timing.Proc. Natl. Acad. Sci. USA 107: 139–144.

Hatton, K.S., Dhar, V., Brown, E.H., Iqbal, M.A., Stuart, S., Didamo,V.T., and Schildkraut, C.L. (1988). Replication program of activeand inactive multigene families in mammalian cells. Mol. Cell. Biol.8: 2149–2158.

Hetzel, J., Duttke, S.H., Benner, C., and Chory, J. (2016). NascentRNA sequencing reveals distinct features in plant transcription.Proc. Natl. Acad. Sci. USA 113: 12316–12321.

Hiratani, I., Ryba, T., Itoh, M., Yokochi, T., Schwaiger, M., Chang,C.W., Lyou, Y., Townes, T.M., Schubeler, D., and Gilbert, D.M.(2008). Global reorganization of replication domains during embry-onic stem cell differentiation. PLoS Biol. 6: e245.

Hiratani, I., et al. (2010). Genome-wide dynamics of replication timingrevealed by in vitro models of mouse embryogenesis. Genome Res.20: 155–169.

Hirsch, C.N., et al. (2016). Draft assembly of elite inbred line PH207provides insights into genomic and transcriptome diversity in maize.Plant Cell 28: 2700–2714.

Hollister, J.D., and Gaut, B.S. (2009). Epigenetic silencing of trans-posable elements: A trade-off between reduced transposition anddeleterious effects on neighboring gene expression. Genome Res.19: 1419–1428.

Hyrien, O. (2016). Up and down the slope: replication timing and forkdirectionality gradients in eukaryotic genomes. In The Initiation ofDNA Replication in Eukaryotes, D.L. Kaplan, ed (Switzerland:Springer International Publishing), pp. 65–85.

Jackson, D.A., and Pombo, A. (1998). Replicon clusters are stableunits of chromosome structure: evidence that nuclear organizationcontributes to the efficient activation and propagation of S phase inhuman cells. J. Cell Biol. 140: 1285–1295.

Jiao, Y., et al. (2017). Improved maize reference genome with single-molecule technologies. Nature 546: 524–527.

Julienne, H., Zoufir, A., Audit, B., and Arneodo, A. (2013). Humangenome replication proceeds through four chromatin states. PLOSComput. Biol. 9: e1003233.

Kaplan, T., Liu, C.L., Erkmann, J.A., Holik, J., Grunstein, M.,Kaufman, P.D., Friedman, N., and Rando, O.J. (2008). Cell cy-cle- and chaperone-mediated regulation of H3K56ac incorporationin yeast. PLoS Genet. 4: e1000270.

Karnani, N., Taylor, C., Malhotra, A., and Dutta, A. (2007). Pan-Sreplication patterns and chromosomal domains defined by genome-tiling arrays of ENCODE genomic areas. Genome Res. 17: 865–876.

Kersey, P.J., et al. (2016). Ensembl Genomes 2016: more genomes,more complexity. Nucleic Acids Res. 44: D574–D580.

Kidd, A.D., Francis, D., and Bennett, M.D. (1987). Replicon size,mean rate of DNA-replication and the duration of the cell-cycle andits component phases in eight monocotyledonous species of con-trasting DNA C values. Ann. Bot. (Lond.) 59: 603–609.

Kim, S.M., Dubey, D.D., and Huberman, J.A. (2003). Early-replicatingheterochromatin. Genes Dev. 17: 330–335.

Kimura, S., Ishibashi, T., Hatanaka, M., Sakakibara, Y., Hashimoto, J.,and Sakaguchi, K. (2000). Molecular cloning and characterization of

2146 The Plant Cell

a plant homologue of the origin recognition complex 1 (ORC1). Plant Sci.158: 33–39.

Klein, K.N., and Gilbert, D.M. (2016). Epigenetic vs. sequence-dependent control of eukaryotic replication timing. In The Initiationof DNA Replication in Eukaryotes, D.L. Kaplan, ed (Switzerland:Springer International Publishing), pp. 39–63.

Kotogany, E., Dudits, D., Horvath, G.V., and Ayaydin, F. (2010). Arapid and robust assay for detection of S-phase cell cycle pro-gression in plant cells and tissues by using ethynyl deoxyuridine.Plant Methods 6: 5.

Lamb, J.C., Meyer, J.M., Corcoran, B., Kato, A., Han, F., andBirchler, J.A. (2007). Distinct chromosomal distributions of highlyrepetitive sequences in maize. Chromosome Res. 15: 33–49.

Lander, E.S., et al. (2001). Initial sequencing and analysis of the hu-man genome. Nature 409: 860–921.

Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009).Ultrafast and memory-efficient alignment of short DNA sequencesto the human genome. Genome Biol. 10: R25.

Laurent, L.C., et al. (2011). Dynamic changes in the copy number ofpluripotency and cell proliferation genes in human ESCs and iPSCsduring reprogramming and time in culture. Cell Stem Cell 8: 106–118.

Lee, M. (1988). The chromosomal basis of somaclonal variation. An-nu. Rev. Plant Physiol. 39: 413–437.

Lee, T.J., et al. (2010). Arabidopsis thaliana chromosome 4 replicatesin two phases that correlate with chromatin state. PLoS Genet. 6:e1000982.

Li, H. (2013). Aligning sequence reads, clone sequences and as-sembly contigs with BWA-MEM. arXiv doi/10.1101/1303.3997.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N.,Marth, G., Abecasis, G., and Durbin, R.; 1000 Genome ProjectData Processing Subgroup (2009). The Sequence Alignment/Mapformat and SAMtools. Bioinformatics 25: 2078–2079.

Lima-de-Faria, A., and Jaworska, H. (1968). Late DNA synthesis inheterochromatin. Nature 217: 138–142.

Liu, R., Vitte, C., Ma, J., Mahama, A.A., Dhliwayo, T., Lee, M., andBennetzen, J.L. (2007). A GeneTrek analysis of the maize genome.Proc. Natl. Acad. Sci. USA 104: 11844–11849.

Lu, F., et al. (2015). High-resolution genetic mapping of maize pan-genome sequence anchors. Nat. Commun. 6: 6914.

MacAlpine, D.M., Rodriguez, H.K., and Bell, S.P. (2004). Co-ordination of replication and transcription along a Drosophilachromosome. Genes Dev. 18: 3094–3105.

Maitra, A., et al. (2005). Genomic alterations in cultured human em-bryonic stem cells. Nat. Genet. 37: 1099–1103.

Makarevitch, I., Eichten, S.R., Briskine, R., Waters, A.J.,Danilevskaya, O.N., Meeley, R.B., Myers, C.L., Vaughn, M.W.,and Springer, N.M. (2013). Genomic distribution of maize faculta-tive heterochromatin marked by trimethylation of H3K27. Plant Cell25: 780–793.

Masai, H., Matsumoto, S., You, Z., Yoshizawa-Sugata, N., andOda, M. (2010). Eukaryotic chromosome DNA replication: where,when, and how? Annu. Rev. Biochem. 79: 89–130.

Masuda, H.P., Ramos, G.B., de Almeida-Engler, J., Cabral, L.M.,Coqueiro, V.M., Macrini, C.M., Ferreira, P.C., and Hemerly, A.S.(2004). Genome based identification and analysis of the pre-replicative complex of Arabidopsis thaliana. FEBS Lett. 574: 192–202.

Masumoto, H., Hawke, D., Kobayashi, R., and Verreault, A. (2005).A role for cell-cycle-regulated histone H3 lysine 56 acetylation in theDNA damage response. Nature 436: 294–298.

Mayshar, Y., Ben-David, U., Lavon, N., Biancotti, J.C., Yakir, B.,Clark, A.T., Plath, K., Lowry, W.E., and Benvenisty, N. (2010).Identification and classification of chromosomal aberrations in hu-man induced pluripotent stem cells. Cell Stem Cell 7: 521–531.

Mechali, M., Yoshida, K., Coulombe, P., and Pasero, P. (2013).Genetic and epigenetic determinants of DNA replication origins,position and activation. Curr. Opin. Genet. Dev. 23: 124–131.

Merchant, N., Lyons, E., Goff, S., Vaughn, M., Ware, D., Micklos,D., and Antin, P. (2016). The iPlant collaborative: cyberinfras-tructure for enabling data to discovery for the life sciences. PLoSBiol. 14: e1002342.

Mickelson-Young, L., Wear, E., Mulvaney, P., Lee, T.J., Szymanski,E.S., Allen, G., Hanley-Bowdoin, L., and Thompson, W. (2016). Aflow cytometric method for estimating S-phase duration in plants.J. Exp. Bot. 67: 6077–6087.

Mori, Y., Yamamoto, T., Sakaguchi, N., Ishibashi, T., Furukawa, T.,Kadota, Y., Kuchitsu, K., Hashimoto, J., Kimura, S., andSakaguchi, K. (2005). Characterization of the origin recognitioncomplex (ORC) from a higher plant, rice (Oryza sativa L.). Gene 353:23–30.

Mroczek, R.J., and Dawe, R.K. (2003). Distribution of retroelementsin centromeres and neocentromeres of maize. Genetics 165: 809–819.

Mukhopadhyay, R., et al. (2014). Allele-specific genome-wide pro-filing in human primary erythroblasts reveal replication programorganization. PLoS Genet. 10: e1004319.

Panning, M.M., and Gilbert, D.M. (2005). Spatio-temporal organiza-tion of DNA replication in murine embryonic stem, primary, andimmortalized cells. J. Cell. Biochem. 95: 74–82.

Peacock, W.J., Dennis, E.S., Rhoades, M.M., and Pryor, A.J. (1981).Highly repeated DNA sequence limited to knob heterochromatin inmaize. Proc. Natl. Acad. Sci. USA 78: 4490–4494.

Percival, D.B., and Walden, A.T. (2000). Wavelet Methods for TimeSeries Analysis. (Cambridge, UK: Cambridge University Press).

Phillips, R.L., Kaeppler, S.M., and Olhoft, P. (1994). Genetic in-stability of plant tissue cultures: breakdown of normal controls.Proc. Natl. Acad. Sci. USA 91: 5222–5226.

Picard, F., Cadoret, J.C., Audit, B., Arneodo, A., Alberti, A., Battail,C., Duret, L., and Prioleau, M.N. (2014). The spatiotemporal pro-gram of DNA replication is associated with specific combinations ofchromatin marks in human cells. PLoS Genet. 10: e1004282.

Plohl, M., Luchetti, A., Mestrovic, N., and Mantovani, B. (2008).Satellite DNAs between selfishness and functionality: structure,genomics and evolution of tandem repeats in centromeric (hetero)chromatin.Gene 409: 72–82.

Pope, B.D., et al. (2014). Topologically associating domains are sta-ble units of replication-timing regulation. Nature 515: 402–405.

Pryor, A., Faulkner, K., Rhoades, M.M., and Peacock, W.J. (1980).Asynchronous replication of heterochromatin in maize. Proc. Natl.Acad. Sci. USA 77: 6705–6709.

Quinlan, A.R., and Hall, I.M. (2010). BEDTools: a flexible suite ofutilities for comparing genomic features. Bioinformatics 26: 841–842.

R Development Core Team (2016). R: A Language and Environmentfor Statistical Computing. (Vienna, Austria: R Foundation for Sta-tistical Computing).

Rafalski, A., and Morgante, M. (2004). Corn and humans: re-combination and linkage disequilibrium in two genomes of similarsize. Trends Genet. 20: 103–111.

Ramos, G.B., de Almeida Engler, J., Ferreira, P.C., and Hemerly,A.S. (2001). DNA replication in plants: characterization of a cdc6homologue from Arabidopsis thaliana. J. Exp. Bot. 52: 2239–2240.

Regulski, M., et al. (2013). The maize methylome influences mRNAsplice sites and reveals widespread paramutation-like switchesguided by small RNA. Genome Res. 23: 1651–1662.

Rhind, N., and Gilbert, D.M. (2013). DNA replication timing. ColdSpring Harb. Perspect. Biol. 5: a010132.

DNA Replication Timing Program in Maize 2147

Rivera-Mulia, J.C., and Gilbert, D.M. (2016). Replicating large ge-nomes: divide and conquer. Mol. Cell 62: 756–765.

Rivera-Mulia, J.C., et al. (2015). Dynamic changes in replicationtiming and gene expression during lineage specification of humanpluripotent stem cells. Genome Res. 25: 1091–1103.

Rivin, C.J., Cullis, C.A., and Walbot, V. (1986). Evaluating quantita-tive variation in the genome of Zea mays. Genetics 113: 1009–1019.

Robinson, J.T., Thorvaldsdottir, H., Winckler, W., Guttman, M.,Lander, E.S., Getz, G., and Mesirov, J.P. (2011). Integrative ge-nomics viewer. Nat. Biotechnol. 29: 24–26.

Rodgers-Melnick, E., Vera, D.L., Bass, H.W., and Buckler, E.S.(2016). Open chromatin reveals the functional maize genome. Proc.Natl. Acad. Sci. USA 113: E3177–E3184.

Rodriguez, J., and Tsukiyama, T. (2013). ATR-like kinase Mec1 fa-cilitates both chromatin accessibility at DNA replication forks and rep-lication fork progression during replication stress. Genes Dev. 27: 74–86.

Roudier, F., et al. (2011). Integrative epigenomic mapping definesfour main chromatin states in Arabidopsis. EMBO J. 30: 1928–1938.

Ryba, T., Battaglia, D., Pope, B.D., Hiratani, I., and Gilbert, D.M.(2011). Genome-scale analysis of replication timing: from bench tobioinformatics. Nat. Protoc. 6: 870–895.

Ryba, T., Hiratani, I., Lu, J., Itoh, M., Kulik, M., Zhang, J., Schulz,T.C., Robins, A.J., Dalton, S., and Gilbert, D.M. (2010). Evolu-tionarily conserved replication timing profiles predict long-rangechromatin interactions and distinguish closely related cell types.Genome Res. 20: 761–770.

Salic, A., and Mitchison, T.J. (2008). A chemical method for fast andsensitive detection of DNA synthesis in vivo. Proc. Natl. Acad. Sci.USA 105: 2415–2420.

Samaniego, R., de la Torre, C., and de la Espina, S.M.D. (2002).Dynamics of replication foci and nuclear matrix during S phase inAllium cepa L. cells. Planta 215: 195–204.

SanMiguel, P., and Vitte, C. (2009). The LTR-Retrotransposons ofMaize. In Handbook of Maize: Genetics and Genomics, J.L.Bennetzen and S. Hake, eds (New York: Springer), pp. 307–327.

SanMiguel, P., et al. (1996). Nested retrotransposons in the inter-genic regions of the maize genome. Science 274: 765–768.

Savadel, S.D., and Bass, H.W. (2017). Take a look at plant DNAreplication: Recent insights and new questions. Plant Signal. Behav.12: e1311437.

Schnable, P.S., et al. (2009). The B73 maize genome: complexity,diversity, and dynamics. Science 326: 1112–1115.

Schwaiger, M., Stadler, M.B., Bell, O., Kohler, H., Oakeley, E.J.,and Schubeler, D. (2009). Chromatin state marks cell-type- andgender-specific replication of the Drosophila genome. Genes Dev.23: 589–601.

Serrano, M., and Blasco, M.A. (2001). Putting the stress on senes-cence. Curr. Opin. Cell Biol. 13: 748–753.

Shultz, R.W., Tatineni, V.M., Hanley-Bowdoin, L., and Thompson,W.F. (2007). Genome-wide analysis of the core DNA replicationmachinery in the higher plants Arabidopsis and rice. Plant Physiol.144: 1697–1714.

Takebayashi, S., Ryba, T., and Gilbert, D.M. (2012). Developmentalcontrol of replication timing defines a new breed of chromosomaldomains with a novel mechanism of chromatin unfolding. Nucleus3: 500–507.

Tanurdzic, M., Vaughn, M.W., Jiang, H., Lee, T.J., Slotkin, R.K.,Sosinski, B., Thompson, W.F., Doerge, R.W., and Martienssen,R.A. (2008). Epigenomic consequences of immortalized plant cellsuspension culture. PLoS Biol. 6: 2880–2895.

Thurman, R.E., Day, N., Noble, W.S., and Stamatoyannopoulos,J.A. (2007). Identification of higher-order functional domains in thehuman ENCODE regions. Genome Res. 17: 917–927.

Trapnell, C., Pachter, L., and Salzberg, S.L. (2009). TopHat: dis-covering splice junctions with RNA-Seq. Bioinformatics 25: 1105–1111.

Trapnell, C., Williams, B.A., Pertea, G., Mortazavi, A., Kwan, G.,van Baren, M.J., Salzberg, S.L., Wold, B.J., and Pachter, L.(2010). Transcript assembly and quantification by RNA-Seq revealsunannotated transcripts and isoform switching during cell differ-entiation. Nat. Biotechnol. 28: 511–515.

Tucker, S., Vitins, A., and Pikaard, C.S. (2010). Nucleolar dominanceand ribosomal RNA gene silencing. Curr. Opin. Cell Biol. 22: 351–356.

Van’t Hof, J. (1976). Replicon size and rate of fork movement in earlyS of higher plant cells (Pisum sativum). Exp. Cell Res. 103: 395–403.

Van’t Hof, J. (1996). DNA Replication in Plants. (Cold Spring Harbor,NY: Cold Spring Harbor Laboratory Press).

Van’t Hof, J., and Bjerknes, C.A. (1981). Similar replicon propertiesof higher plant cells with different S periods and genome sizes. Exp.Cell Res. 136: 461–465.

Van’t Hof, J., Bjerknes, C., and Clinton, J. (1978a). Replicon prop-erties of chromosomal DNA fibers and the duration of DNA syn-thesis of sunflower root-tip meristem cells at different temperatures.Chromosoma 66: 161–171.

Van’t Hof, J., Kuniyuki, A., and Bjerknes, C. (1978b). Size andnumber of replicon families of chromosomal DNA of Arabidopsisthaliana. Chromosoma 68: 269–285.

Vera, D.L., Madzima, T.F., Labonne, J.D., Alam, M.P., Hoffman,G.G., Girimurugan, S.B., Zhang, J., McGinnis, K.M., Dennis, J.H.,and Bass, H.W. (2014). Differential nuclease sensitivity profiling ofchromatin reveals biochemical footprints coupled to gene expressionand functional DNA elements in maize. Plant Cell 26: 3883–3893.

Versteeg, R., van Schaik, B.D., van Batenburg, M.F., Roos, M.,Monajemi, R., Caron, H., Bussemaker, H.J., and van Kampen,A.H. (2003). The human transcriptome map reveals extremes ingene density, intron length, GC content, and repeat pattern fordomains of highly and weakly expressed genes. Genome Res. 13:1998–2004.

Vonholdt, B.M., Takuno, S., and Gaut, B.S. (2012). Recent retro-transposon insertions are methylated and phylogenetically clus-tered in japonica rice (Oryza sativa spp. japonica). Mol. Biol. Evol.29: 3193–3203.

Wang, B., Tseng, E., Regulski, M., Clark, T.A., Hon, T., Jiao, Y., Lu,Z., Olson, A., Stein, J.C., and Ware, D. (2016). Unveiling thecomplexity of the maize transcriptome by single-molecule long-read sequencing. Nat. Commun. 7: 11708.

Wang, X.F., Elling, A.A., Li, X.Y., Li, N., Peng, Z.Y., He, G.M., Sun,H., Qi, Y.J., Liu, X.S., and Deng, X.W. (2009). Genome-wide andorgan-specific landscapes of epigenetic modifications and theirrelationships to mRNA and small RNA transcriptomes in maize.Plant Cell 21: 1053–1069.

Wear, E.E., Concia, L., Brooks, A.M., Markham, E.A., Lee, T.-J.,Allen, G.C., Thompson, W.F., and Hanley-Bowdoin, L. (2016).Isolation of plant nuclei at defined cell cycle stages using EdU la-beling and flow cytometry. Methods Mol. Biol. 1370: 69–86.

West, P.T., Li, Q., Ji, L., Eichten, S.R., Song, J., Vaughn, M.W.,Schmitz, R.J., and Springer, N.M. (2014). Genomic distribution ofH3K9me2 and DNA methylation in a maize genome. PLoS One 9:e105267.

Williams, S.K., Truong, D., and Tyler, J.K. (2008). Acetylation in theglobular core of histone H3 on lysine-56 promotes chromatin dis-assembly during transcriptional activation. Proc. Natl. Acad. Sci.USA 105: 9000–9005.

Witmer, X., Alvarez-Venegas, R., San-Miguel, P., Danilevskaya, O.,and Avramova, Z. (2003). Putative subunits of the maize origin ofreplication recognition complex ZmORC1-ZmORC5. Nucleic AcidsRes. 31: 619–628.

2148 The Plant Cell

Wolfgruber, T.K., et al. (2009). Maize centromere structure andevolution: sequence analysis of centromeres 2 and 5 reveals dy-namic loci shaped primarily by retrotransposons. PLoS Genet. 5:e1000743.

Xu, F., Zhang, K., and Grunstein, M. (2005). Acetylation in histone H3globular domain regulates gene expression in yeast. Cell 121: 375–385.

Yan, S., Zhang, Q., Li, Y., Huang, Y., Zhao, L., Tan, J., He, S., and Li,L. (2014). Comparison of chromatin epigenetic modification pat-terns among root meristem, elongation and maturation zones inmaize (Zea mays L.). Cytogenet. Genome Res. 143: 179–188.

Yang, S.C., Rhind, N., and Bechhoefer, J. (2010). Modeling genome-wide replication kinetics reveals a mechanism for regulation ofreplication timing. Mol. Syst. Biol. 6: 404.

Zhang, M., et al. (2014). Genome-wide high resolution parental-specific DNA and histone methylation maps uncover patterns ofimprinting regulation in maize. Genome Res. 24: 167–176.

Zhang, Y., et al. (2008). Model-based analysis of ChIP-Seq (MACS).Genome Biol. 9: R137.

Zhao, H.N., Zhu, X.B., Wang, K., Gent, J.I., Zhang, W.L., Dawe,R.K., and Jiang, J.M. (2016). Gene expression and chromatinmodifications associated with maize centromeres. G3 (Bethesda) 6:183–192.

Zhong, C.X., Marshall, J.B., Topp, C., Mroczek, R., Kato, A.,Nagaki, K., Birchler, J.A., Jiang, J., and Dawe, R.K. (2002).Centromeric retroelements and satellites interact with maize kinet-ochore protein CENH3. Plant Cell 14: 2825–2836.

Zink, D. (2006). The temporal program of DNA replication: new in-sights into old questions. Chromosoma 115: 273–287.

Zynda, G.J., Song, J., Concia, L., Wear, E.E., Hanley-Bowdoin,L., Thompson, W.F., and Vaughn, M.W. (2017). Repliscan: a toolfor classifying replication timing regions. BMC Bioinformatics 18:362.

DNA Replication Timing Program in Maize 2149

DOI 10.1105/tpc.17.00037; originally published online August 25, 2017; 2017;29;2126-2149Plant Cell

Matthew W. Vaughn, Linda Hanley-Bowdoin and William F. ThompsonLorenzo Concia, Patrick Mulvaney, Eric S. Szymanski, George C. Allen, Robert A. Martienssen,

Emily E. Wear, Jawon Song, Gregory J. Zynda, Chantal LeBlanc, Tae-Jin Lee, Leigh Mickelson-Young,) Root Tipsmays

ZeaGenomic Analysis of the DNA Replication Timing Program during Mitotic S Phase in Maize (

 This information is current as of March 20, 2020

 

Supplemental Data /content/suppl/2017/08/25/tpc.17.00037.DC1.html

References /content/29/9/2126.full.html#ref-list-1

This article cites 141 articles, 54 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists