growing an embryo from a single cell: a hurdle in animal...

Growing an Embryo from a Single Cell:A Hurdle in Animal Life

Patrick H. O’Farrell

Department of Biochemistry, University of California San Francisco, San Francisco, California 94158

Correspondence: [email protected]

A requirement that an animal be able to feed to grow constrains how a cell can grow into ananimal, and it forces an alternation between growth (increase in mass) and proliferation(increase in cell number). A growth-only phase that transforms a stem cell of ordinaryproportions into a huge cell, the oocyte, requires dramatic adaptations to help a nucleusdirect a 105-fold expansion of cytoplasmic volume. Proliferation without growth transformsthe huge egg into an embryo while still accommodating an impotent nucleus overwhelmedby the voluminous cytoplasm. This growth program characterizes animals that deposit theireggs externally, but it is changed in mammals and in endoparasites. In these organisms,development in a nutritive environment releases the growth constraint, but growth of cellsbefore gastrulation requires a new program to sustain pluripotency during this growth.

The phrase “Nothing in biology makes senseexcept in the light of evolution,” originally

introduced in an essay supporting the theoryof evolution (Dobzhansky 1973), has been re-purposed to chide biologists, who all too oftenignore the origin and purpose of the biologi-cal processes and mechanisms they study. Tolook for how these processes “make sense,” oneshould look at the way they serve the life historiesof the organisms in which they are used. Such anexamination can suggest how a process mighthave been molded by evolution to achieve thebenefit provided. I have been an advocate forsuch considerations, because I believe them tobe the best guide to interesting and importantquestions for investigation. Here, I will synthe-size observations made by examination of thenatural histories of diverse organisms to de-

scribe four almost universal phases of growthand proliferation in animal biology. I will befocusing on the two earliest phases: the growththat produces a huge egg and the transformationof this single large cell into an embryo. The sub-sequent phases of growth were described in pre-vious reviews (O’Farrell 2004, 2011; O’Farrellet al. 2004). But first, I will provide some contextfor the comparisons that are made, and a briefsynopsis of the four growth phases that convertan ordinary-sized stem cell into a big metazoanadult.

MAMMALS CAME LATE

It is perhaps important to point out to the read-er that I will first focus on animals that deposittheir eggs externally, as this oviparous lifestyle

Editors: Rebecca Heald, Iswar K. Hariharan, and David B. Wake

Additional Perspectives on Size Control in Biology: From Organelles to Organisms available at www.cshperspectives.org

Copyright # 2015 Cold Spring Harbor Laboratory Press; all rights reserved

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a019042

1

on March 13, 2020 - Published by Cold Spring Harbor Laboratory Press http://cshperspectives.cshlp.org/Downloaded from

mailto:[email protected]



http://www.cshperspectives.org



http://cshperspectives.cshlp.org/

is the most general and ancestral mode of ani-mal development. Mammals, of course, are in-teresting to us, but if we wish to understandour connection to the rest of biology, we shouldfirst recognize the context within which mam-mals appeared. Mammals are a small clade of�4000 species out of more than two millionanimal species, and they are derived chordatesthat arose relatively late in animal evolution.Furthermore, mammalian evolution investedin a special program, which, by housing embry-os in a nutritive environment, introduced anunusual fashion of dealing with the growth con-straints faced by all embryos. Understanding thespecializations of mammalian development inthis context can be illuminating (O’Farrell et al.2004). I will end with an effort to point outthese mammalian specializations and deep con-nections to programs showed by their evolu-tionary predecessors. But we will begin with afew guiding generalizations that give perspectiveon the coordination of growth with animal de-velopment.

AN ORGANISM MUST DEVELOP FEEDINGSTRUCTURES BEFORE IT CAN INCREASEIN MASS

Although there are some exceptionally largesingle-celled organisms, large body plans aresubstantially the domain of multicellular organ-isms. However, the production of a large bodyfrom a single-cell zygote must deal with a fun-damental problem. Any animal whose nutritiondepends on a complex body plan with its spe-cialized feeding structures must be able to de-velop these specialized structures before it canfeed and grow independently. In these cases, acomplex body plan needs to be produced dur-ing development before there is significantgrowth of the organism (O’Farrell 2004).

We will view life histories of organisms inthe light of this problem, which appears to haveacted as a constraint throughout the evolutionof the animals. Despite great diversity, this van-tage point reveals relationships that lead to theglobal concept of four phases of growth andproliferation in the life plans of animal species.

ALIGNING PROGRAMS OF DEVELOPMENTAT A CONSERVED STAGE

The strategy used when comparing distantly re-lated protein sequences suggests a general ap-proach when looking for distant homologies.Rather than just aligning sequences at the aminoterminus, one first identifies the most conserveddomains and uses these for alignment. Similarly,in comparing the programs of growth and pro-liferation of diverse species, I have chosen toalign the life histories of different organisms ata particularly conserved point (O’Farrell 2004;O’Farrell et al. 2004).

The early developmental biologists, von Baerand Haeckel, were fascinated by the similaritiesof vertebrate embryos of very diverse organismsat a stage following establishment of the bodyaxis and formation of the neural tube. Recogni-tion of this conserved stage and description ofthe morphogenetic events that brought increas-ingly dramatic distinctions to the embryos ofdifferent species were taken by some as evidencefor a claim that ontogeny recapitulates phyloge-ny—or that the developmental sequence passesthrough all of the evolutionary steps as if “moreadvanced” organisms achieve their distinctionsby late additions to the developmental sequence(e.g., Graham and Richardson 2012). The attrac-tive phrase, “ontogeny recapitulates phylogeny,”appears to be somewhat of an overstatement andthe idea is often described as discredited. None-theless, the existence of a relatively similar em-bryonic morphology is strongly supported.

A common embryonic morphology is alsofound among embryos within different inverte-brate phyla. This is particularly apparent in ar-thropods. Arthropod embryos of very diversespecies show remarkably similar morphologieson the initial establishment of the body plan.These similar looking embryos also use similarmolecules to guide the formation of embryonicpatterns (Patel et al. 1989; Prud’homme et al.2003). This point in development has beencalled the phylotypic stage to reflect the factthat diverse organisms within a phylum show a“typical morphology” at this stage.

It is notable that the “phylotypic” stages ofdifferent phyla occur at a similar point during

P.H. O’Farrell

2 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a019042



embryogenesis, just after establishment of thebasic body plan, and embryos of even differentphyla have a very similar organization at thisstage. These distant similarities suggested bymorphologies have been supported by molecu-lar analyses (Domazet-Loso and Tautz 2010; Ka-linka et al. 2010). Key mechanisms patterningthe phylotypic embryo are conserved across phy-la. For example, a gradient in a signaling mole-cule of the transforming growth factor (TGF)-btype guides dorsoventral polarity of the embryo,and local expression of different homeotic-typetranscription factors subdivide the anteroposte-rior axis of the body plan (Shen 2007). Thesefindings suggest mechanistic parallels and arguefor evolution by descent. That is, even distantlyrelated animals evolved from a common prede-cessor with a similar embryonic stage, which waspatterned by mechanisms that continue to beused in extant animals. Consequently, the phy-lotypic stage can be taken as an especially con-served stage of development at which the embry-os of different species can be compared.

The original view exemplified by the “ontog-eny recapitulates phylogeny” phrase presumesthat embryogenesisbegins with acommon ances-tral morphology that adds species-specific spe-cializations as embryogenesis progresses. Howev-er, the phylotypic stage occurs after importantearly steps in embryogenesis and the modernview considers development more like an hour-glass (Raff 1996; Prud’homme and Gompel2010). Eggs are diverse, and distinctive programsand morphologies are seen during very early em-bryogenesis, but diversity shrinks with progres-sion to the phylotypic stage and then reexpandsin later development. The bottleneck in morpho-logical diversity at the phylotypic stage presum-ably is a reflection of the early evolution of a suc-cessful strategy for buildinga basic body plan,andthe developmental reexpansion of diversity afterthe phylotypic stage is a likely an indication thatmodification and addition of specializations weremajor routes of later evolutionary diversification.

SUBDIVIDING ANIMAL GROWTH

Embryos at the phylotypic stage are not onlysimilar in general morphology, but they are the

same size, or relatively close (O’Farrell 2004).Taking a little license with the exact stage thatis defined as the phylotypic stage, the anterior/posterior axis of the phylotypic embryo isroughly a millimeter. A few examples are a littlesmaller (e.g., Caenorhabditis elegans) or bigger(e.g., grasshopper), and some embryos continueto extend their length for some time after theinitial elaboration of the body axis. Nonetheless,sizes are rather similar. This similarity in sizeapplies to the embryos of fruit flies, frogs, andblue whales, organisms whose adult masses,1 mg, 100 g, and 100 metric tons, range over1011-fold. This near-constancy in the size ofthe embryo at the phylotypic stage allows us todivide our consideration of growth in animalsinto two. How do embryos of �10 mg (wetweight of cytoplasm) at the phylotypic stagegrow to the adult sizes? How does oogenesisand early embryogenesis transform a single or-dinarily sized stem cell into a phylotypic em-bryo, a miniorganism with a roughed-out bodyplan?

In this presentation, I consider growth asan exponential process. Thus, the seeminglymassive growth of a 7-ton blue whale calf intoa 150-ton adult is viewed as �20-fold and,hence, considerably less impressive than the�100,000-fold growth of a normally sizedstem cell to produce a 10-mg Drosophila egg.

Examination of the life histories of numer-ous species reveals four dramatically differentphases of growth in the life plans of many ani-mals:

1. Sponsored growth: During oogenesis, a stemcell grows enormously in size without divi-sion. The mother sponsors this growth forthe benefit of the progeny.

2. Subdivision: Following deposition of an eggin an external environment, the massive eggdivides rapidly without any growth to pro-duce the phylotypic embryo.

3. Expansion: Diverse but identifiable strate-gies are used in the growth of the phylotypicembryo to the adult.

4. Size control: Adult organisms usually con-strain their growth, a process that relies on

Growing an Embryo from a Single Cell

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a019042 3



both quiescence and balanced replacementstrategies.

The central features of growth in phases 3and 4 were described in two previous reviews(O’Farrell 2004, 2011). Together with the con-siderations presented here, these reviews reveala perspective gained from examination of thenatural histories of organisms and their devel-opment and growth. It is strikingly differentfrom the views we develop when focusing onsome of our models of cell growth and prolifer-ation. I suggest that the biological perspectivegained in an exercise of considering how growthand proliferation are controlled in animal de-velopment will uncover new and important fac-ets of the control of these processes.

PHASE 1—SPONSORED GROWTH:GROWING A HUGE CELL (OOGENESIS)

General Background

The development of an externally deposited egginto a feeding organism depends on suppliesreceived from the mother. This maternal dowryhas two parts: a large volume of cytoplasm, of-ten in the range of 100,000 times the amountfound in a normal cell, and a supply of nutri-ents, largely in the form of yolk. The first growthphase is maternally sponsored growth of a stemcell of ordinary proportions into a large oocytethat is often more than 105-fold bigger than anaverage cell. This phase represents growth with-out division and it involves highly specializedprocesses that are well studied in only a fewsystems.

The vast majority of the millions of animalspecies deposit their eggs in the external envi-ronment in which, without further nutritionalinput, they develop within protective shells orcoating that largely isolate the embryo from itsenvironment. The duration and extent of thedevelopment that occurs during this autono-mous stage varies substantially in proportionto the nutritional supplies from the mother.From species to species, the amount of yolkincluded in an egg differs over many orders ofmagnitude. Avian eggs are at the upper end ofthis spectrum. For example, a chicken egg has a

yolk vesicle of �17 g that is actually one extra-ordinarily large cell, the oocyte, before fertiliza-tion. Imagine a dinosaur oocyte! The chickenoocyte is more than a hundred-thousand-million-fold (1011) larger than a typical cell,which is �5 mm in diameter (or �10210 cm3,here approximated as 10210 g). On a geometricscale, the 1011-fold growth to produce this largecell in an ovary of the hen totally belittles theroughly 100-fold growth of the hatchling to anadult chicken. This growth of the single-celledoocyte is, of course, growth without prolifera-tion. Although the growth of the yolk-rich avianoocyte is above average, and so exaggerates thecontribution of oogenesis to overall growth, it isnonetheless true that, on a geometric scale, thegrowth of an oocyte makes huge contributionsto the growth of organisms throughout animalphylogeny.

Yolk versus Cytoplasm

Before embarking on a consideration of thephases of growth, I would like to deconstructthe extravagant process that occurs in the hen.Besides familiarity, I use the chicken egg as anexample because of the dramatic size of the oo-cyte. As mentioned, the size of the yolk suppliesprovided by the mother differs widely betweenspecies, and it is the extravagant supply of yolkthat makes the avian oocyte a standout in size.But chicken oocytes also have a very large cyto-plasm, and here eggs of even very diverse organ-isms are remarkably similar. In the case of thechicken oocyte, the nonyolky cytoplasm formsbut a small whitish disc, the blastodisc, on thesurface of the huge yolk. Although small on arelative scale, this small white area includes�10 mg of cytoplasm, a cytoplasmic volumeof 100,000 times that of an ordinary cell. Thegrowth that produces this volume of cytoplasmamounts to a remarkable achievement for onecell, and I consider this growth as the first of thefour phases of growth.

After fertilization, this egg cytoplasm willdivide rapidly to generate a cap of cells on topof the massive yolk vesicle, and these cells willundergo morphogenesis to produce the ele-mentary body plan of the phylotypic embryo.

P.H. O’Farrell




This subdivision of cytoplasm and productionof a phylotypic embryo represents phase 2 of thegrowth and proliferation program, the subdivi-sion phase that is largely attributed to prolifer-ation without growth. For the moment, what Iwant to get across is that yolk contributes littleto this process. To a large extent, it is the non-yolky cytoplasm that undergoes this phase 2program to produce an elemental phylotypicembryo, and later the yolk is called on to funda third phase in the growth and proliferationprogram, the expansion phase (O’Farrell 2004,2011; O’Farrell et al. 2004). Thus, we will ignoreyolk in our considerations of phase 1 and phase2 growths. We will instead focus on the cyto-plasmic accumulation in oogenesis and its sub-division in early embryogenesis.

Transcriptional Capacity, a Constraint onProduction of Large Cells

There are numerous examples of huge cells: theDrosophila salivary gland cells, the cells of Asca-ris (a nematode that has a volume a billion-foldbigger than C. elegans but the same cellular anat-omy), the giant cells of Aplasia, and ciliates ofextraordinary size. But these cells have to eitherovercome or tolerate a biological constraint. Thegenome has a limited ability to support growthor rapid changes in a large cell. As pointed out byWoodland (1982), the rate at which a gene canproduce transcripts is limited, and this limita-tion imposes practical limitations on the growthand function of large cells. Let us look at thislimitation.

Rates of initiation of transcription cannotbe increased indefinitely. There is a packing lim-it on the density of RNA polymerase moleculesthat can be loaded along a DNA molecule. Onceinitiation rates are high enough to load a se-quence to this limit, no further increase in therate of RNA synthesis can be achieved. Irrespec-tive of the length of a gene, at the maximumpacking density, polymerases are lined up onebehind the other at spacing estimated to be 50–100 bp. A new transcript will be produced onlywhen a polymerase advances to the end of thegene, and the rate at which each successive po-lymerase arrives at the end—corresponding to

the rate of product synthesis—depends only onthe rate at which the polymerase traverses thedistance separating it from the previous poly-merase. If we use 30 bp per sec as an approxi-mation of transcriptional elongation rate and60 bp as the maximal packing density, a newpolymerase will advance to the end of the geneevery 2 sec. Thus, at maximal loading, a single-copy gene can produce a new transcript onceevery 2 sec.

A normally sized animal cell has �100,000messenger RNAs. The most abundantly ex-pressed genes in most cells represent several per-cent of the total and would be present at levelsbetween �1000 and 10,000 copies. At the max-imal rate of gene expression, it would take be-tween 2000 and 20,000 sec, or between 30 minand 5 h, to make this amount of transcript, andhalf this time in a diploid and half again in a G2

cell, which has four copies of the genome. Agrowing cell must be able to produce a new com-plement of all of its mRNAs every time it dou-bles. For many animal cells, this requirement isnot a serious constraint because doubling timesare usually considerably longer than the timerequired to produce the most abundant tran-scripts. However, as a cell enlarges, the amountof mRNA in the cell increases in rough propor-tion to the cytoplasmic volume. Many oocytesachieve a size �100,000 times that of a normalcell. It would take comparably longer to makethe transcripts to supply this large cytoplasm.Above, we calculated that a normal cell wouldtake 0.5 to 5 h to make its most abundant tran-scripts. If we increase the lower value 100,000-fold, we find that it would take �50,000 hor �2000 d. For a G2 cell, with its four copies ofeach gene, this reduces to 500 d. According tothis, the growth of cells to a very large size shouldbe an extremely slow process, and take roughly ayear.

If an organism produces exceptionally largecells, it must do so very slowly or find a solutionto the transcriptional rate constraint. In manycases, and in all of the examples of large cellsgiven above other than oocytes, the genome isamplified to increase coding potential. In ac-cord with the idea that transcriptional potentialmight limit growth, manipulations of the ploi-





dy of Drosophila salivary gland cells have shownthat the size attained depends on ploidy (Fol-lette et al. 1998; Hayashi and Yamaguchi 1999).But an increase in ploidy does not appear to bean option for the germline, as this would dis-rupt the normal and faithful transmission ofchromosomes from generation to generation.Ciliated protozoa have found a way to enjoythe benefits of polyploidy while retaining stablegenetics. Here, a disposable polyploid macro-nucleus supports large cells, while a distinctdiploid germ nucleus, the micronucleus, bearsthe responsibility for inheritance. How do oo-cytes in animals manage to grow despite thelimited transcriptional potency of their nuclei?

Nurse Cell Support of Oocyte Growth

One effective way to speed the growth of a singlelarge cell is to enlist the help of other cells. Aspecial organization is required if other cells areto contribute transcripts to a growing oocyte. Inmany species of arthropods and annelids, theprecursor of the oocyte divides, but only incom-pletely, leaving sister cells interconnected by cy-toplasmic bridges that develop into specializedand stable intercellular conduits called ring ca-nals. The interconnected group of sister cells iscalled a cyst when it first forms, and, in insects inwhich it is a stable unit in oogenesis, it is calledan egg chamber once it is invested in a coating ofindependently derived follicle cells. Althoughone cell of the interconnected group becomesthe oocyte, the remaining interconnected cellsserve as nurse cells that supply the oocyte. Thenurse cells amplify their genomes and their ca-pacity to produce RNA in endoreduplication cy-cles consisting of distinct rounds of S phasewith-out cell division. The resulting large polyploidcells efficiently fuel the growth of an immenseegg cytoplasm. Forexample, in Drosophila, a pre-cursorcell called acystocyte divides four times tocreate an interconnected “cyst” of 16 cells thatdevelops into an egg chamber in which 15 ofthe cells become highly polyploidy, and onecell, which will grow into the oocyte, arrests inmeiotic prophase. Although the oocyte nucleusremains transcriptionally inactive, the nursecells produce an abundance of RNA and cyto-

plasmic constituents that are transferred to theoocyte, which grows enormously over a few daysat the expense of the nurse cells.

The nurse cell strategy is particularly effec-tive in overcoming the limitation imposed bytranscriptional capacity and it supports the rap-id increase in cytoplasmic volume of the devel-oping oocyte. The mother can also make othercontributions to the development of the egg.Most notably, a different approach is generallyused to accelerate the accumulation of the yolkynutrient supply of the oocyte. Other tissues ofthe mother, even distant tissues, such as liver,can contribute yolk protein, vitellogenin, lipids,and other nutrients that are taken up by the oo-cyte often with the assistance of surroundingfollicle cells. The accumulation of yolk is of ma-jor importance to nutrition of the embryo andlater growth. The mother also provides furthercontributions in the elaboration of protectivecoatings and shells. But here we focus on theroughly 100,000-fold growth of the cytoplasm,exclusive of yolk, as the germline stem cell ex-pands to produce the oocyte. The numerouscopies of the genome available in nurse cells,altogether a few thousand copies in the eggchamber of Drosophila melanogaster, provideabundant template for rapid accumulation oftranscripts and growth of the oocyte. In organ-isms that use this strategy, a stem cell can matureinto a large oocyte relatively quickly, �10 d inthe case of Drosophila.

Autonomous Growth of Oocytes

Cytology and electron microscopy have beenused to survey oogenesis in numerous organ-isms. These analyses define different oogenesisstrategies. In many species, the precursors of oo-cytes either lack or lose connections with sistercells and the individual oocytes grow autono-mously, often with an isolating coating of folli-cle cells that do not have detectable cytoplasmiccontinuity with the growing oocyte. In these or-ganisms, the oocyte nucleus appears to supportthe growth in cytoplasmic volume, and the con-tributions of the maternal environment appearto be limited to provision of yolk and other nu-trients and production of protective coatings.

P.H. O’Farrell




Given the limitation imposed by the tran-scriptional potency of a single nucleus, howdoes autonomous growth succeed at producingoocytes? To a large degree, the answer is thatoocytes using this program grow much moreslowly, generally taking several months to morethan a year to reach a mature size (Smiley 1990).But even this pace appears to rely on severalstriking specializations that appear designed toenhance provision of key transcripts.

As expected, when growth is autonomous,the oocyte nucleus is transcriptionally active(Gall et al. 2004). The growth phase still occursduring an arrested prophase of meiosis I, butmeiosis is modified to incorporate a specialtranscriptional phase generally not seen in sys-tems using nurse-cell-supported growth. Meio-sis begins normally. The chromatids condenseon the start of prophase, and progress throughthe meiotic events of homolog pairing and re-combination before a diplotene arrest. At thisarrest, large loops of chromatin unfurl from astill-visible chromosome axis. The axes outlinethe meiotic arrangement of paired chromo-somes, suggesting a change in chromatin com-paction without disruption of the meiotic chro-mosome configuration. The chromatin loopsare packed with RNA polymerase and elongatingtranscripts giving a striking appearance noted byearly microscopists. Because they resembled thelong fuzzy brushes that were once used to cleanthe soot from the glass chimneys on oil-burninglamps, they are called lampbrush chromosomes.This stage often persists for months as the oocyteenlarges, and studies in Xenopus showed thatoocyte transcripts accumulate during this peri-od (Gall 2012; Kloc et al. 2014).

The lampbrush chromosome stage providesa long period of highly active transcription witha postreplicative nucleus that has four copies ofthe genome. Still, our rough calculations wouldsuggest that a very long time would be requiredfor the transcript accumulation needed for a100,000-fold increase in cytoplasmic volume.It would not be surprising if there were addi-tional adaptations that might reduce the severityof the transcriptional constraint so that oocytescould be produced more rapidly. Indeed, there isdramatic evidence for such adaptations.

Selective Amplification of Genes EncodingRibosomal RNA

Unlike protein-coding RNAs whose impactis further amplified by translation, structuralRNAs, such as ribosomal RNA, function di-rectly, and there is no way to bypass or minimizethe need for a high transcriptional capacity toaccumulate a high level of functional product.Furthermore, there are a lot of ribosomes in evenan ordinary cell, roughly several million (Blobeland Potter 1967; Wolf and Schlessinger 1977), afew orders of magnitude more than most abun-dant mRNAs. If there were only one gene encod-ing ribosomal RNA, the nucleus would takeabout a month to meet the transcriptional de-mand required to reproduce a normally sizedcell. However, ribosomal RNA is encoded in re-peating arrays of genes at loci often called nucle-olar organizers. The repeats increase the tran-scriptional capacity, but phenotypes resultingfrom partial deletions of the recombinant DNA(rDNA) repeats in Drosophila suggest that, evenwith this repetition, there is little reserve capac-ity (Gersh 1968). In species in which oocytesgrow autonomously, the coding capacity ofthe rDNA repeats would be severely taxed andshould limit growth. However, a remarkableprocess takes care of this limitation.

As oocytes progress through meiotic pro-phase to their arrest in diplotene, they hugelyamplify the ribosomal repeats (Brown and Da-wid 1968; Gall 1968). The replication associatedwith this amplification has been detected bylabeling, and the DNA copy increase was as-sessed by molecular and cytological methods.This amplification can be so extensive that itincreases the total amount of DNA in the nu-cleus several-fold. The amplified sequences areproduced as free circles that form hundreds ofsmall nucleoli throughout the oocyte nucleus.Amplification occurs in diverse organisms thatuse oocyte-autonomous growth and is pre-sumed to support the massive increase in ribo-somes during growth.

We have seen three striking biologicalinnovations that promote growth of oocytes:nurturing by nurse cells, transcription fromlampbrush chromosomes, and amplification





of rDNA. Apparently, the nurse-cell strategy re-duces the need for other adaptations to increasetranscriptional output, as oocytes supported bynurse cells appear to be transcriptionally silentand to lack lampbrush chromosomes (Ganotet al. 2008). These relatively widespread special-izations modify three fundamental biologi-cal processes. First, cytokinesis is rendered in-complete and the arrested cleavage furrows aretransformed into ring canals that interconnectnurse cells and oocytes. Second, although tran-scription is normally discontinued on mitoticor meiotic chromosomes, the lampbrush chro-mosome stage is created by reactivating tran-scription of condensed meiotic chromosomes.Finally, although DNA replication is ordinarilytightly regulated to ensure that all sequences arereplicated once and only once, these controlsare bypassed to massively amplify the rDNArepeats. These are not oddities of a peculiarbranch of evolution—they are modificationsmaking key contributions to a central step inthe life history of nearly all animal species, theproduction of an oocyte.

Unbalanced Growth

The accepted truism about cell growth is that allof the contents of a cell need to double, at leaston average, every time a cell divides. This appliesto the much-studied case of exponentially grow-ing cells in culture. However, growth to producean organism is not simply a matter of increasingeverything equally. Growth in vivo is often un-balanced, and, at a cellular level, this is particu-larly dramatic when one considers the hugegrowth of a stem cell into an oocyte. Extraordi-nary changes occur in the size of some struc-tures, and the stoichiometry of some key cellularcomponents. As I describe here, this places hugestress on growth itself as a result of dispropor-tionate changes in pathways contributing togrowth.

The nucleus of the oocyte is huge—so bigthat it has its own special name, the germinalvesicle. Despite having only the DNA content ofa normal G2 cell, the nucleus expands in roughproportion to the entire oocyte, �100,000times. The DNA occupies only a tiny volume

of the germinal vesicle, and forms an easilystained compact structure called the karyosomewithin a sea of nucleoplasm. Avariety of special-ized structures have been described in the ger-minal vesicle (Gall et al. 2004; Gardner et al.2012). One of these structures, the mininucleoli,which appear throughout the germinal vesicle,results from the amplification of ribosomal re-peats and is clearly a response to the dispropor-tionate demand for ribosomal RNA synthesisdescribed above.

The growth/multiplication of mitochon-dria in the oocyte faces huge distortions in thebalance of different mitochondria gene prod-ucts, those encoded in the nucleus and thoseencoded by the mitochondrial DNA. The mito-chondrial genome is replicated in proportionto the expansion of the cytoplasm. In Xenopus,7 � 108 copies of mitochondrial DNA are pro-duced during oogenesis (Chase and Dawid1972), whereas the nuclear DNA is unchanged.This represents nearly a million-fold increase inmitochondrial genome copy number above thatin a typical cell. Because nuclear and mitochon-drial DNA–encoded gene products assemblestoichiometically to make the respiratory com-plexes, continued production of functional mi-tochondria must be managed in the face of achanging imbalance in the two contributions.We know nothing of the mechanisms that dealwith this supply imbalance, one that is faced to asmaller degree in other enlarging cells, such asneurons.

The great bulk of the multiplication of mi-tochondria occurs in a specialized structure, theBalbiani body. This structure, which is found inthe oocytes of a wide range of animals, is seen asa differentiated area of cytoplasm that can bedetected early in oogenesis, just as precursorcells are committing to oogenesis (Kloc et al.2014). It is usually adjacent to the nucleus. Of-ten, it is a dominating structure obvious in thelight microscope. In the electron microscope,it is seen as a congregation of mitochondria,endoplasmic reticulum, and an RNA-rich struc-ture called nuage (French for “cloud”). Nuageappears to be universally associated with de-veloping oocytes and more generally withthe germline. Localization, genetic dissection

P.H. O’Farrell




and identification of molecular constituentshave tied nuage not to one, but to several func-tions—storing mRNAs for the future embryo,specifying the vegetal or posterior pole ofthe egg, defining the germline, and protectingagainst infectious DNA elements by housing animmunity system based on a category of smallRNAs, piRNAs (Marlow and Mullins 2008; Klocet al. 2014). Obviously, there are many tasks to beperformed in the oocyte, and rapid progress isbeing made in defining the contributions ofnuage. But here, we focus on the mitochondria.

The cytoplasmic Balbiani body of Xenopus,which is densely packed with mitochondria,grows to be huge. Its increase in size is paralleledby an increase in mitochondria and mitochon-drial DNA, and the structure only dispersesmonths later when the oocyte turns its attentionto yolk accumulation. Mitochondrial morphol-ogy and DNA labeling argue that Balbiani bodymitochondria proliferate (Chase and Dawid1972; Webb and Smith 1977; Kloc et al. 2014).Indeed, it appears that more than 16 doublingsof the mitochondrial genome occur before dis-persal of the Balbiani body in Xenopus.

The Balbiani body is not always a dominat-ing morphological feature. In Drosophila, it isdifficult to see by light microscopy without spe-cialized tags, and while obvious by electron mi-croscopy, it is rather transient in comparison toits persistence in autonomously growing oo-cytes (Cox and Spradling 2003). This is likelylinked to delegation of some responsibility forproduction of various components to the mul-tiple nurse cells. In earwigs, an insect wherein asingle large nurse cell supports each oocyte, aBalbiani body is found in the nurse cell as well asthe oocyte (Tworzydlo et al. 2009).

Nurse-cell support of oogenesis, as exempli-fied by Drosophila, reduces one type of stress.The increased nuclear transcription capacityprovided by the polytene nurse cells reducesthe imbalance in nuclear and mitochondrialcopy number. However, the transcription of nu-clear products occurs in a distinct, albeit inter-connected, nurse cell compartment. Now, thereis a need to manage the flow of gene product tothe oocyte where mitochondria are actively rep-licating. Perhaps more impressive is that most of

the 100,000-fold increase in the number of mi-tochondrial genomes occurs in a few days in thismore rapid style of oogenesis.

One final feature of mitochondrial multipli-cation dramatically emphasizes the fact thatgrowth in vivo is “unbalanced”—that not allcellular components double when cells double.In C. elegans, a supply of �100,000 mitochon-drial genomes is accumulated during oogenesis,and no further increase occurs until nearly adultstages. Additionally, a mutant lacking the lateincrease in mitochondrial genomes producesmature adults (Tsang and Lemire 2002; Braticet al. 2009). Thus, the supply of genomes madeduring oogenesis suffices for the life of theworm. It is likely that the behavior seen in C.elegans is broadly representative, although likelyless extreme in other systems. After the hugeincreases in mitochondrial DNA copy numberduring oogenesis in Xenopus and in Drosophila,little or no further increase occurs during all ofembryogenesis (Chase and Dawid 1972; H Maand PH O’Farrell, unpubl.).

Although these descriptions of mitochon-drial multiplication during oogenesis do not re-veal mechanism, they give avery different view ofwhat mechanisms we should be looking for. Mi-tochondrial genome replication is uncoupledfrom cellular growth and proliferation, but isstrikingly regulated during development andshows a remarkable departure from balancedgrowth. To understand how the copy numberof this genome is regulated, we need to look fordevelopmental inputs, and we might also expectspecial regulatory circuits that handle the rapidmultiplication in the face of a dramaticallychanging context during oocyte growth.

Constrained Change versus Conservation

The specialized features described above can befound in diverse organisms with representativesin more than one phylum. For example, lamp-brush chromosomes have been described in or-ganisms representing mollusks, arthropods,echinoderms, and chordates (Bedford 1966;Gruzova and Batalova 1979; Smiley 1990; Gallet al. 2004). This gives the impression of conser-vation, and conservation often comes with a





connotation of stability. However, change is per-sistent, and it imposes a constant diversifyingpressure that is resisted by selection. Constraintsimposed by a requirement for function limitsome change, but not all. Thus, like the carsproduced over the past several decades, whichcontinue to have four wheels (generally), asteering mechanism and, brakes, a lot of changesoccur “under the hood.” Variations in the for-mat of oogenesis and the distributions of thevariants in phylogeny reveal changes, but theretained features suggest operation of powerfuland lasting constraints that limited the range ofsuccessful formats.

Although our sampling of species is still rel-atively small, the phylogenetic distribution ofnurse-cell-supported oogenesis versus autono-mous oogenesis is of interest. Nurse-cell-sup-ported oogenesis is seen in numerous insect spe-cies and has been particularly well studied inDrosophila; however, there are also many insectsincluding members of one of the most diversegroups of animals on earth, the beetles, whichshow autonomous growth of their oocytes. Be-cause the nurse-cell format of oogenesis isfound, at least so far, in derived insect lineages,it is suspected that this program itself is derived.This interpretation is consistent with findings ofautonomous oocyte growth in sampled speciesof various other animal phyla, including chor-dates, echinoderms, mollusks, and cnidarians,and a group of more primitive animals, such asjellyfish (Smiley 1990; Eckelbarger and Larson1992; Eckelbarger and Young 1997; Marlow andMullins 2008). However, the nurse-cell programis not limited to derived insect lineages. Nursecells are also used in annelids, and a relatednurse-cell type of oogenesis has been discoveredin a group of chordate species whose lifestylerelies on rapid oogenesis (Ganot et al. 2007).Furthermore, a group of frogs that carry theireggs in a brooding pouch, marsupial frogs,does oogenesis differently than other frogs, andthe part that is especially different has featureslike nurse-cell-supported oocyte growth. Inthese frogs, oogonia undergo multiple incom-plete divisions to create a dramatically multinu-cleate oocyte: these nuclei develop lampbrushchromosomes, apparently all contributing tran-

scripts for growth of the oocyte, but eventuallyall but one germ nucleus is eliminated (DelPino and Humphries 1978; Macgregor and DelPino 1982). Also, nematodes produce eggs froma syncytial gonad in which many nuclei contrib-ute to a cytoplasm that is partitioned into therapidly formed large egg (O’Farrell 2004). Thus,helpercells, orhelper nuclei, promote thegrowthof large oocytes by means of shared cytoplasmin at least a few animal phyla. This might repre-sent convergent evolution to this strategy.

An alternative to convergent evolution isdivergence from an unrecognized common pro-genitor. Indeed, at least one observation mightsuggest a deep connection between nurse-cell-supported oogenesis and autonomous oogene-sis. In numerous organisms, cytokinesis of theprecursors to the oocyte is incomplete and pro-duces specialized bridges interconnecting thesister cells (Pepling et al. 1999; Marlow and Mul-lins 2008). In organisms with nurse-cell pro-grams, only one cell of the interconnected groupbecomes the oocyte, and the bridges persist sothat the other cells, the nurse cells, can transmittheir cytoplasmic endowment to the oocyte.However, in organisms showing autonomousoocyte growth, the relationship is transient, islost before oocyte growth, and usually each ofthe originally connected cells makes an oocyte.Although a common progenitor to nurse-cellprograms and autonomous oogenesis has notbeen recognized, we should not forget that sexappeared earlier in evolution than animals. Per-haps vestiges of early gametogenesis programsprovided a foundation for the different strate-gies present in animal species today.

Regardless of the path taken in evolution,representative animals in every major branchof phylogeny grow large oocytes, and these var-ied organisms call on a limited set of specializedmechanisms to support this growth. I suggestthat this reflects the persistent operation of con-straints that guide selection. An operationalnecessity to produce an egg that can supportdevelopment to an autonomous feeding stagemight have been the driver of other constraintsas follows. Development of a feeding organismwithout growth demanded a large egg. Produc-tion of a large egg required extensive growth of

P.H. O’Farrell




the oocyte. Production of very large cytoplasmharboring only a single nucleus distorted theusual biological relationships, and the resultingstresses produced selection for special adapta-tions.

PHASE 2—FROM EGG TO EMBRYO

Fertilization and egg deposition marks a majorturnaround for growth. After growing to hugedimensions as a single cell, nutrients are no lon-ger freely available as the egg sets out to make anembryo. In doing so, the huge egg is subdividedinto many cells and is restructured, initiallywithout growth. It is not a time to tarry. Theegg is a defenseless immobile package of nutri-ents, and speed of development is one way thatexternally deposited eggs minimize predation.

We tend to think of development as a rela-tively protracted affair, perhaps because of ourown gestation time. But the long periods of fetalgrowth in mammals are more about growth thandevelopment, as is the time of yolk-supportedgrowth of birds. The relationship between devel-opmental time and growth is perhaps most ob-vious in birds, in which the time to hatching indifferent species increases in proportion to thesize of the egg and the hatchling it eventuallyproduces (Rahn et al. 1974). Externally depos-ited eggs with little yolk often develop quickly.C. elegans develops from a fertilized egg into afunctional organism in 13 h. Some clam specieshatch as ciliated trochophore larvae within 24 h,a Drosophila larva crawls away from its eggshell24 h after fertilization, and zebrafish swim after3 d. Apparently, the extraordinary processesof morphogenesis and differentiation that buildan organism do not need to take a long time,but growth does. To sidestep variations associ-ated with growth, we will consider developmentfrom fertilization to the time of the phylotypicembryo as the second phase of “growth” andproliferation. I have called this phase “subdivi-sion,” because it primarily involves subdivisionof the mass of the oocyte, without growth (i.e.,without a substantial increase in mass).

As we saw for phase 1 growth, the limitedtranscriptional potency of a nucleus also con-strains early developmental events during phase

2. Slow accumulation of RNA was a chief limi-tation to the growth of a huge oocyte, and nowthe egg has this same cumbersomely large cyto-plasm and a single nucleus. Slow accumulationof transcripts would stymie the progress of de-velopment (Woodland 1982). As a result, eggsinitially avoid relying on transcription.

The egg immediately embarks on a programthat will resolve the discordance between thecytoplasmic volume and the nucleus. Rapidcell cycles exponentially increase the numberof nuclei, whereas the cytoplasmic volume re-mains constant and so reduces the cytoplasmfor which each nucleus is responsible. As thenuclear to cytoplasmic ratio increases, tran-scription can, and does, begin to play an impor-tant role in regulation.

Importantly, the early cell cycles that rampup the transcriptional potential of the embryoare themselves independent of transcription.Regulated gene expression normally plays a ma-jor role in the control of a whole host of biolog-ical processes, including regulation of the cellcycle, and, by forsaking transcriptional control,the embryo has to extensively restructure bio-logical regulatory mechanisms. Consequently,the egg, at the starting block of development,uses unusual biological mechanisms. As I willdiscuss below, the unusual aspects of this ances-tral program of fast, transcription-independentcell cycles has a pervasive impact on the biologyof the early embryo and influences the ground-work on which subsequent development hasbeen built. But first, I will very briefly reviewmore studied aspects of this early embryonicprogram.

Rapid Cell Cycles and the MidblastulaTransition (MBT)

The rapid early embryonic divisions have beenan attractive model for cell-cycle studies, andthe return to reliance on transcription hasbeen an interesting area of developmental anal-ysis. Fast, early cell cycles are widespread.The mitotic cycles of newly fertilized eggs ofDrosophila, sea urchin, and Xenopus are 8.6,30, and 30 min, respectively. These cycles lackgap phases, and occur even if transcription is





prevented—indeed, in frog and sea urchin, theyhave been shown to occur even if the nucleus iseliminated (Harvey 1935). After several period-ic divisions, the cycles begin to slow, at firstgradually and then more abruptly, as the em-bryo approaches gastrulation.

In addition to its transcriptional indepen-dence, the early cell cycle is redesigned in severalways to achieve speed. It lacks the gap phases, G1

and G2, which usually separate S phase and mi-tosis. Shortening of S phase by as much as 100-fold further quickens these cycles (Farrell andO’Farrell 2014). These features of the early cellcycle are seen in diverse settings—the rapidlydividing teleoblasts of annelid embryos, thesyncytial mitotic cycles of Drosophila, the earlydivisions of sea urchin, and those of Xenopus.Thus, these cycles, which are often viewed as pe-culiar and distinct from “normal,” are used dur-ing the initial steps of development of diverseanimals. I suggest that the constancy of thesefeatures is the result of constraints demandingrapid expansion of transcriptional potential inthe unwieldy large-egg cytoplasm.

One of the most fascinating questions re-garding these early cycles is: What terminatesthem? The egg is primed by its maternal dowryto initiate relentless and rapid cycles that haveno obvious outside input. But the rapid cycleshave to end, and they slow in a stereotyped fash-ion just before gastrulation. Furthermore, thematernally deposited gene products need to re-linquish control to transcription of the zygoticgenome. This changeover, which involves de-struction of maternally provided RNAs andproteins and activation of zygotic transcription,is referred to as the maternal-to-zygotic transi-tion (MZT). In part, this is a progressive tran-sition with additional zygotic functions cominginto play as persisting maternal functions aresuccessively eliminated (Wieschaus 1996; Fol-lette and O’Farrell 1997). Nonetheless, duringthis process, there are abrupt transitions whenthe nearly silent nuclei activate expression ofnew panels of genes (Newport and Kirschner1982a,b; Edgar and Schubiger 1986; Edgar etal. 1986). The slowing of the cell cycle, MZT,and onset of the cell-shape changes and move-ments that initiate morphogenesis occur over a

relatively short time span that is called the mid-blastula transition (MBT) (Farrell and O’Farrell2014).

Although it is not yet understood how theMBT is controlled, it seems natural to believethat the embryo is running out of something—perhaps a key component is used up, or a ma-ternal supply becomes inadequate as nuclei in-crease. Experimental manipulation of the ratioof nuclei to cytoplasm suggests that its increasehas an input into the MBT (Newport andKirschner 1982a,b; Edgar and Schubiger 1986;Edgar et al. 1986; Pritchard and Schubiger 1996).Progress has been made in understanding howthe cell cycle slows. For example, inhibitoryphosphorylation leads to a decline in activityof the cell-cycle kinase, Cdk1, prolongs S phase,and introduces a G2 phase at the MBT in Dro-sophila (Edgar and O’Farrell 1990; Farrell andO’Farrell 2013). Increasing nuclear density ap-pears to one of the signals triggering slowing ofthe cell cycles; however, the mechanism of thiscoupling remains unclear. It is also uncertainwhether the nuclear to cytoplasmic ratio actsdirectly on other MBT events, or whether itseffects are mediated by changes in the cell cycle.

Distinctive Regulation during Early Cell Cycles

Beyond suggesting that growth issues underlie auniversal program of early cell cycles withoutgrowth or transcription, creation of this specialproliferation stage has been associated with amajor retooling of regulation so that embryonicdevelopment begins in a special context.

In addition to impressive diversity in na-ture, evolution creates distinctions between tis-sues and between stages in the life histories of anorganism. Exemplifying the latter, evolutionoptimizes larval and adult forms for differentlifestyles. Phase 2 of animal growth is present inthe life histories of all animals, and evolutionmight have optimized this stage. As evidence forsuch an evolutionary trajectory, I present threeaspects of regulatory biology that are widelyconserved among eukaryotes, and yet are notfollowed by early embryos.

Cells coordinate the synthesis of the precur-sors of DNA, the deoxynucleotide triphosphates

P.H. O’Farrell




(dNTPs), with replication. By providing thedNTPs just as they are used, their concentrationnever rises very high, and hydroxyurea inhibi-tion of ribonucleotide reductase, an enzyme keyto producing dNTPs, quickly brings DNA rep-lication to a standstill in organisms from bacte-ria to human. Things are different in the earlyembryo. Xenopus eggs have high levels of dNTPs(Woodland and Pestell 1972). Additionally,hydroxyurea does not block the early S phasesof Xenopus or of Drosophila (Landstrom et al.1975; Newport and Dasso 1989; Howe et al.1995; K Yuan, AW Shermoen, PH O’Farrell, un-publ.). Sea urchin eggs initiate cleavage cycleswithout one of the subunits of ribonucleotidereductase and must similarly have a preformedpool of dNTPs (Evans et al. 1983). Thus, eggs inat least three different phyla deviate from anotherwise universal program of regulation andcarry a pool of dNTPs that provides DNA pre-cursors for all of the early cell cycles.

Histones assemble newly replicated DNAinto chromatin, and, like the dNTPs, histonesare usually made only during S phase. Again ear-ly embryos are different. The egg carries a stock-pile of histone proteins. The egg also has a largecapacity to store this depot in complex withchaperones, such as nucleoplasmin (Laskey etal. 1978). This deviation from an otherwise uni-versal program regulating histone supply couldhave a major consequence on the chromatin as-sembled.

Two types of chromatin exist in eukaryotesfrom yeast to humans: heterochromatin andeuchromatin. The especially compacted het-erochromatin, originally defined by intense his-tologic staining, is now more commonly dis-tinguished by molecular hallmarks (e.g., ahistone modification, H3-K9 methylation, andbinding of heterochromatin protein 1 [HP1]).These marks are considered repressive becauseheterochromatin is generally transcriptionallyquiescent. Additionally, heterochromatic DNAreplicates later than euchromatic sequences,and the sequential replication of different re-gions of the genome is the main reasons that Sphase is normally long (Shermoen et al. 2010).But during the early cell cycles, chromatin isrelatively decompacted, repressive marks are

absent, and all sequences replicate at the sametime (Shermoen et al. 2010; Li et al. 2014; Yuanet al. 2014). Thus, the specialization of chroma-tin into euchromatic and heterochromatic do-mains, a feature of regulatory biology that spansknown eukaryotic biology, is abandoned in theearly embryonic cell cycles.

It is dangerous to suggest why things are theway they are, but it seems likely that the dramaticdistinctions in mechanisms that characterizethe early embryo are secondary to the devotedattention to the pell-mell cell cycles. Cell-cycleevents can disrupt other processes. The spindleusurps cytoskeletal components, and cytokine-sis disrupts intercellular membranes. Further-more, nascent transcripts are aborted duringmitosis, interrupting ongoing transcription un-til new RNA polymerases transit the entire gene(Shermoen and O’Farrell 1991). In the earlyembryonic cycles, cells are in mitosis for a largefraction of the cycle, and many cell biologicalprocesses appear to be held in abeyance.

In summary, constraints forced an alterna-tion between different styles of growth. Eachstyle of growth elicited evolutionary adapta-tions that altered fundamental features of bio-logical regulation. Consequently, the transitionsfrom one growth phase to another are widelytransformative.

Erasing the Slate for Development

Because animals evolved with alternating growthphases as a consistent context, developmentalprocesses might be optimized to function intheir phase-specific biological setting. In thisway, the unusual biology of the proliferation-only phase might have become an integral partof programs governing early animal develop-ment. Here I consider pluripotency—the abilityof cells to form multiple cell and tissue types.

Induction of pluripotent stem cells in mam-mals is highly topical as result of current ex-citement about stem cells and their therapeuticpotential. However, the capacity of a cell to pro-duce diverse cell types is broadly important. Asdevelopment progresses and cells differentiate,their genomes and the chromatin that packagesthese genomes are modified to restrict gene





expression. But all life forms must begin eachgeneration anew with a totipotent progenitor.

Though the epigenetic marks that specifythe fates of cells can be transmitted over manycell divisions, John Gurdon’s classical experi-ments showed that a nucleus of a differentiatedcell, when transplanted into an enucleated frogegg, could support development of a viable andfertile frog (Gurdon et al. 1958). Thus, the egg,acting either directly or by driving the implant-ed nucleus through a rapid series of maternallyprogrammed divisions, can expand the devel-opmental potency of an otherwise inadequatenucleus.

The resetting of epigenetic marks is a com-plex topic as there are many types of marks andcomplex interactions influencing their produc-tion, inheritance, and reversal. It is now knownthat pluripotency can be established by experi-mental treatments, but the egg cytoplasm is spe-cial in that it is both capable of resetting epige-netic marks and is the natural environment inwhich such a resetting should occur. To get aview of how the egg might reprogram a nucleus,I consider its specialized ability to activate oneof the most repressed and condensed nucleiknown, that of the sperm.

Following fertilization, the highly con-densed sperm nucleus swells and the chromatindecompacts, replicates, and forms the male pro-nucleus, which joins the female pronucleus toparticipate in the rapid early-division cycles. Inthe early stages of this process, sperm nuclearbasic proteins (SNBPs), which include special-ized histones and protamines, are replaced withmore conventional histones. Xenopus egg ex-tracts promote this decompaction of sperm nu-clei. Immunodepletion showed that this activitydepends on nucleoplasmin, the abundant his-tone chaperone (Philpott et al. 1991). Nucleo-plasmin-like chaperones in flies, sea urchins,and frogs appear to be involved in sperm de-compaction.

The octomeric nucleosome is built in a two-step process by deposition of H3/H4 tetramerfollowed by addition of a H2A/H2B tetramer.Nucleoplasmin interacts with H2A/H2B, andso appears specialized for the second step. Afly mutant, hira, was unable to decompact the

sperm nucleus, which then failed to contributeto zygotic events (Loppin et al. 2005). Hira is aconserved member of a protein complex thatacts as an H3/H4 chaperone to promote repli-cation-independent deposition of nucleosomes(Orsi et al. 2013). The Hira homologs in fishand mouse are similarly required for decompac-tion (Zhao et al. 2011; Lin et al. 2014).

Decompaction is a general activity of eggextracts with extracts of one species being ableto decompact the sperm of various species, de-spite differences in the SNBPs. Furthermore,these extracts will decompact the chromatin ofother nuclei, such as the inactive nuclei of avianred blood cells. I suggest that the ability to ex-change basic chromatin proteins, whether prot-amines or histones, with a pool of naıve histonescontributes to the ability of the egg to repro-gram the epigenetic state of a nucleus (see Linet al. 2014).

In contrast to egg extracts, oocyte extracts areineffective at sperm decompaction. Nucleoplas-min is extensively phosphorylated at the time ofegg activation, and dephosphorylation of nucle-oplasmin compromises its activity in decompac-tion (Philpott et al. 2000). Phosphorylation ofnucleoplasmin persists until the MBT is depen-dent on high-cyclin-dependent kinase activity, aspecialization of the proliferation-only phase.Additionally, the egg and early cleavage stageembryo are well positioned to reprogram epige-netic marks because they have especially highlevels of the histone chaperones and a uniquepool of naıve histones available for exchange.

The speed of the early embryonic cyclesmight also contribute to reprogramming byoutpacing the mechanisms responsible for epi-genetic inheritance. Histone methylation andDNA methylation marks are added after DNAreplication and histone deposition, and, if cellsprogress immediately to mitosis and anotherround of replication, there may not be adequatetime to modify the newly replicated DNA andnewly deposited histones. Indeed, during theextremely rapid early cell cycle in Drosophilaembryos, chromatin lacks hallmarks of hetero-chromatin, H3K9 methylation, and bound HP1binding, and these only make their appearanceas the cycles slow. When the features of hetero-

P.H. O’Farrell




chromatin do appear, they do so in a sequenceas if heterochromatin formation is an integralpart of the developmental program (Shermoenet al. 2010).

I suggest that erasure of the epigenetic marksin externally deposited eggs is coupled to theunusual context in the egg and early embryo,and that restoration of epigenetic marks awaitsslowing of the cell cycle when chromatin be-comes less dynamic and can again accumulatemodifications. As discussed below, even thoughthese mechanisms are set in a different contextin mammalian development, an understandingof the ancestral mechanism will give us deeperinsight into this process in mammals (Canonet al. 2011; Sanchez-Sanchez et al. 2011; Ma-randel et al. 2012; Lee et al. 2013).

WHY ARE MAMMALS SO DIFFERENT?

Mammalian eggs are smaller than externally de-posited eggs, and their early cleavages are slow.So, despite a long heritage of predecessors thatadhered to the four phases of growth that I havedescribed, the early growth phases of mamma-lian embryos are altered. Here, I explore thenature of the difference, and why the mamma-lian program differs.

The Alignment Problem

One of the most dramatic oddities of mamma-lian development is that it does not start at thesame point as the development of other animals.A developmental module generating extra em-bryonic tissues is inserted at the beginning ofmammalian development. This insertion ob-scures relationships even with close vertebraterelatives (O’Farrell et al. 2004). Certainly, whenI was faced with descriptions that began withfertilization, I was hard pressed to see commonfeatures among of the developmental programsof key vertebrate models—frog, chicken, andmouse.

If, instead, we align development at the phy-lotypic stage (O’Farrell et al. 2004), one is struckby the close relationships among the embryos ofthese model vertebrate organisms, as was vonBaer two centuries ago. Moving from this point

of alignment to earlier stages reveals a point ofdivergence in the programs. In all three organ-isms, gastrulation precedes establishment of thephylotypic body plan and is regulated by relatedmolecules and mechanisms, even if differencesin topology of the movements obscure the par-allel. In frog and chicken, as well as avast numberof organisms throughout phylogeny, the rapidcleavage cycles of the egg immediately precedegastrulation so that the blastomeres produced bycleavage make up the gastrula. In contrast, cleav-age of the mouse egg begins 6.5 d before the startof gastrulation, and the inner cell mass (ICM)cells and trophoblast cells that result from thesecleavages do not initiate gastrulation (Fig. 1),but instead begin a growth program discussedbelow. Recognition of the changes associatedwith the insertion of the mammalian specificmodule gives us new ways of thinking aboutthe parallels between organisms.

Insertion of Mass Increase into Phase 2Growth

Protected and fed by the maternal environ-ment, the mammalian embryo is releasedfrom growth constraint, and its altered devel-opment depends on the relaxation of the con-straint. Mammalian eggs are small, and have�100–1000 times less cytoplasm than a typicalexternally deposited egg. Consequently, mam-malian embryos need to grow to produce aphylotypic embryo of ordinary proportions. Agrowth period added to embryogenesis pro-vides the chance to “catch up” to the embryosof externally deposited eggs.

The early mammalian egg begins develop-ment at a leisurely pace. One to two cell divisionsper day are typical in early mammalian embryos.This slow beginning is a feature that appears inthe derived mammalian lineage and is not seenin closely related vertebrates. For example, in thetime it takes a mouse embryo to reach its seconddivision, a chicken egg has generated a 60,000-cell blastodisc and is preparing to gastrulate. At3 d, the still small mouse embryo has formeda blastocyst with 8 to 10 inner cell mass cells,which will contribute to the embryo proper.But these cells are far from ready to begin devel-





opment at this point. Over the course of the nextday and a half, the blastocyst hatches out of itsprotective shell (zona pellucida) and implants.On hatching, it begins to grow rapidly. Duringthe next 3 d, the embryo grows in volume 500-fold (Snow 1981). There is also an impressiveincrease in cell number arriving at 15,000 cellsat the time of gastrulation. Although this catch-up phase of growth makes a mammalian gastru-la that is comparable in size and cell number tothe gastrulae formed by externally depositedeggs, it also changes the foundation for gastru-lation. It diminishes maternal input into em-bryogenesis and separates the cleavage cyclesfrom the onset of patterning.

To understand the relationship of the mam-malian program to the ancestral program, it isimportant to understand at what point themammalian module has been inserted. I suggestthat this module was inserted near the very be-ginning (see asterisk on the ancestral program inFig. 1). Accordingly, the “cleavage” of the mam-

malian egg is part of the inserted program;hence, it is not surprising that its featuresbear little resemblance to the features of earlydivisions in externally deposited eggs. This ar-rangement suggests that the rapid cleavagecycles of the ancestral program have been dis-placed and are part of the growth program, aposition that seems incongruous for “cleavagecycles,” but we will see that, although there areimportant modifications to the early divisionprogram, this alignment fits numerous obser-vations.

Realignment of Early Development

According to the proposed displacement of ear-ly developmental events to a later stage in mam-mals, we should be able to identify parallels be-tween the proliferative division that amplify cellnumber in the mammalian embryo and thecleavage cell cycle. One parallel is timing withrespect to conserved features of embryogene-

Gastrulation

Ancestral program

Phylotypicembryo

Mammalian program

*

Figure 1. Insertion of a new module of development distinguishes early mammalian embryogenesis from that ofits predecessors. Most animals deposit very large eggs in the external environment where they develop into afeeding organism without nutrition or growth. Cleavage cell cycles divide the egg into blastomeres that imme-diately initiate pattern formation and differentiation to produce an embryo, called phylotypic, because all of thespecies of a phylum look similar at this stage. Even the embryos of other phyla are similar in shape and size at thisstage. Mammalian embryogenesis, as exemplified by the mouse, begins with a much smaller egg (magnifiedsomewhat here) and a distinctly different early program of development. The cells resulting from the initialdivisions do not gastrulate. First, extra embryonic tissues are produced as the embryo grows enormously andrapidly. Only then does it gastrulate and establish strong parallels to the ancestral program of embryogenesis. Themammalian module of development (boxed) is inserted early during development (�) so that the earliestdivisions are dramatically changed and many of the attributes of the early events of the ancestral program aredeferred along with gastrulation.

P.H. O’Farrell




sis—both types of divisions immediately pre-cede gastrulation; another is speed.

The cell cycle just before gastrulation of themammalian embryo are the fastest known inmammals. A 5.1-h cell-cycle time would beneeded to explain the increasing cell number ifgrowth were uniform. However, proliferation isnot uniform, and, based on mitotic index, Snowestimated that cells just anterior to the streak, themorphological marker of gastrulation, doubleevery 2–3 h (Snow 1977). Rat embryos show asimilarly fast cell-cycle rate of �3–3.5 h nearthe streak (Mac Auley et al. 1993). Thus, thesecycles, like the cleavage cycles of externally de-posited eggs, are exceptionally fast. In additionto being fast, other similarities that I have previ-ously reviewed suggest parallels between thesepregastrulation cell cycles in mammals and theearly cleavage cycles (O’Farrell et al. 2004). Here,I present the relationship between mammalianembryonic stem (ES) cells and the insertedgrowth phase, as well as describing apparentparallels of ES cell cycles to early embryonic cy-cles in externally deposited eggs.

ES cells were derived from the ICM of mouseblastocysts (Martin 1981). As Martin suggestedin her report of the successful isolation of EScells, perhaps the cells “represent a selected pop-ulation of completely normal embryonic cellsthat are programmed to divide until they receivethe appropriate signals for differentiation.” In-deed, the ICM cells divide and grow during therapid-growth phase of the implanted mamma-lian embryo, and these cells are still growing atthe egg cylinder stage, which also successfullyyields pluripotent stem cells (Evans and Kauf-man 1981). Thus, growing ES cells provide amodel for the growth phase of the pregastrula-tion mammalian embryo.

ES cells also have a peculiar cell cycle thatreverts to more a typical mode of regulationwhen these cells are induced to differentiate (Sa-vatier et al. 1996). Before differentiation, ES cellsdivide quickly for a cultured cell, although notnearly as fast as the estimates for the embryoniccells. The ES cells also have remarkably short G1

and G2 phases, reminiscent of the absence of G1

and G2 in the cleavage cycles of externally depos-ited eggs (Coronado et al. 2013). Furthermore,

the cycle is regulated unlike other culturedmammalian cells. During G1, the canonical reg-ulators of progress to S phase, Rb, and cyclin E,are inappropriately phosphorylated and abun-dant, respectively (White et al. 2005). Further-more, oscillations of numerous key cell-cycleregulators are greatly dampened, such that thereis peculiarly high cyclin:Cdk activity in inter-phase (Savatier et al. 1996; Fujii-Yamamotoet al. 2005; Yang et al. 2011; van der Laan et al.2013). Although these features of the ES cellcycle have puzzled investigators expecting aninvariant mode of cell-cycle regulation, theyare familiar to those of us exploring the changesin cell-cycle regulation during early develop-ment in other organisms. Notably, work in Dro-sophila has shown that the early rapid cycles areassociated with very high interphase cyclin:Cdkactivity, and near absence of oscillation in theabundance of key cell-cycle regulators (Edgaret al. 1994). Furthermore, there is not just onedivergent mode of regulating the early cell cyclein Drosophila. As development progresses, themode of regulating cell-cycle changes in severalsteps as S phase slows, a G2 is added and then aG1 is introduced. Rb and cyclin E come to playimportant regulatory roles only late in this pro-cess, consistent with their behavior in ES cells. Ipropose that cell-cycle regulation in ES cells, aswell as the embryonic cells from which they arederived, is modified, and these modificationsreflect their derivation from the early embryoniccycles of externally deposited eggs. It will beinteresting to learn the extent of the parallels.

In summary, it appears that the mammalianmodule of development was inserted within theearly rapid cell divisions with modifications tothese divisions. This insertion displaced manydevelopmental events to a later stage.

Pluripotency and the Transfer ofResponsibility to the Zygotic Program

Perhaps the chief modification of the early em-bryonic programs is a switch from maternal pro-graming to zygotic programming of the rapiddivisions in mammals. The maternal program-ming of cleavage divisions used by externallydeposited eggs is not practical in mammals be-





cause the rapid divisions that continue after ex-tensive growth would have diluted maternalgene products. Furthermore, mammals have lit-tle need to relyon maternal programming. Giventhat their eggs are small and that the protectedmammalian egg develops at a leisurely pace, evena single nucleus can modify transcript levels fastenough to regulate events. Indeed, mammalianembryos become dependent on zygotic gene ex-pression within the first few divisions with ki-netics that differ slightly between species. Thisearly zygotic program faces requirements absentin externally deposited eggs. The zygotic pro-gram of mammalian embryos sustains a grow-ing population of pluripotent cells and thenprovides these cells with a zygotic mimic of theancestral maternally run cleavage program.

The addition of the zygotically supportedembryonic growth program in mammalian de-velopment created a new need for a mechanismto sustain pluripotency. In externally depositedeggs, the early rapid divisions that produce theblastomeres appear to be part of the programthat refreshes chromatin structures to create anaıve state for the beginning of development.Because these newly produced blastomeres im-mediately begin morphogenesis and differenti-ation at the time that they initiate zygotic geneexpression, they have no need to maintain theirpluripotency. But the mammalian embryo hasto zygotically create pluripotency and maintainit during a growth phase that amplifies cell num-ber .500-fold before initiating embryogenesis.

Tremendous advances in the study of stemcells have identified mammalian factors that canreprogram differentiated cells to pluripotency(Takahashi and Yamanaka 2006). These pluri-potency factors are expressed in normal embry-onic cells, and sustain pluripotency as the ICMcells of the blastocyst grow and proliferate beforegastrulation (Marandel et al. 2012; Le Bin et al.2014; Sun et al. 2014). If the pluripotent cells ofthe mammalian embryo are analogous to thecleavage stage blastomeres of externally depos-ited eggs, perhaps homologs of the pluripotencyfactors will be expressed in the blastomeres inthe predecessors of mammals. Oct4, Sox2, andNanog, three factors involved in maintenance ofpluripotency in mammals, do have homologs in

several nonmammalian vertebrates, and they areexpressed in early pregastrulation embryos.

As a result of maternal supply and very earlyzygotic expression, the zebrafish homologs ofOct4, Sox2, and Nanog are present in the earlyembryo. They function jointly to promote earlytranscription. In their absence, transcription ofmany genes fails and the embryo does not gas-trulate (Lee et al. 2013). These factors appear toact widely to direct transcription of genes nec-essary to stimulate the developmental potentialof the blastomeres of the zebrafish embryo (Leeet al. 2013; Leichsenring et al. 2013). These find-ings suggest analogies between the pluripotencyof mammalian embryonic cells and transcrip-tion factors directing the onset of the initialzygotic program of gene expression in fish.

Despite relatedness, Oct4, Sox2, and Nanoghave somewhat different jobs in fish and mam-mals. In fish, these transcription factors act as akick-starter to activate the first set of genes thatare transcribed as the embryo transitions to zy-gotic expression. But the initial wave of tran-scription is transient as the embryo immediate-ly advances to gastrulation and differentiationand new tiers of zygotic gene expression. Inmammals, the role of pluripotency factors isless transient. These transcription factors are re-quired to sustain developmental potency duringgrowth and proliferation until the time of thedeferred activation of gastrulation and differen-tiation. Presumably, vertebrates coopted a pre-existing transcription program that drove earlygene expression in vertebrate ancestors, and re-configured the wiring of the program so that theexpression of these early activators, their targets,and the early properties of the cells persist dur-ing the growth phase.

In addition to simple persistence, the pluri-potency factors may have another job. In con-trast to fish eggs, which are preloaded withmaternal products to run the rapid embryoniccycles, the mammalian embryo has to supportits dramatic growth by zygotic gene expression.Perhaps early embryonic transcription factorstook on this growth-promoting role and nowstimulate expression of cell-cycle and growthgenes to drive the rampant pregastrulationgrowth in mammals.

P.H. O’Farrell




The presence of the “pluripotency factors”in cleavage stage zebrafish embryos might sug-gest conservation of the program that erasesepigenetic marks. However, relatively rapid di-vergence of these transcription factors outsideof mammals shows that pluripotency in otherspecies can be accomplished independent ofthese factors. I suggest that the extremely fastmaternally run divisions in externally depositedeggs makes a major contribution to erasing epi-genetic marks, and that the pluripotency factorstook over this role in conjunction with theabandonment of the maternally driven divisionprogram in mammals.

Releasing the Growth Constraint UnleashesEvolutionary Change

I have argued that the simple requirement thatan egg be able to produce a feeding organismgave rise to constraints on the growth programof animals. In particular, it forced the large-eggparadigm, and the requirement for a large egggenerated new constraints that underlie the cu-rious features of growth and proliferation inphases 1 and 2. A viviparous lifestyle opens upnew possibilities. Development of the egg in aprotected uterine environment in which nu-trients can be provided frees evolution of thesegrowth constraints. Once freed of constraints,change is possible and to some extent inevitable.

Several of the major changes seen in earlymammalian development might be explainedby the release of the constraint on growth.Clearly, growth of the embryo following hatch-ing from the zona pellucida depends on therelease of the growth constraint. It also seemslikely that the small size of the mammalian eggand absence of yolk accumulation were allow-able changes because of the release of the growthconstraint. But because many coupled changesoccurred during the evolution of mammals, it ishard to argue that release of the growth con-straint was the driving event. Perhaps other ex-amples of a release of the growth constraint cangive us additional perspective on this.

It turns out that the constraint on growth ofthe embryo has been released numerous timesin evolution, perhaps most frequently in organ-

isms called endoparasites that deposit their eggswithin the bodies of other organisms. This life-style has evolved independently numeroustimes (Grbic et al. 1998; Grbic 2000; Wiegmannet al. 2011), and a diversity of changes in earlydevelopment are found among the species withthis lifestyle (Grbic et al. 1998; Grbic 2000;Wiegmann et al. 2011). Although perhaps notan appealing analogy, the life histories of endo-parasitic wasps provides a number of indepen-dent examples of branches of evolution in whicheggs are released of their responsibility for pro-viding for the development to a feeding adultstage. Additionally, the changes in developmentassociated with these events can be analyzed insome detail because close relatives of the endo-parasitic wasps are known and these adhereclosely to a canonical program of early develop-ment described in detail in Drosophila.

The endoparasitic wasps lay their eggs insidethe larvae of bigger insects where they develop inthe nutrient-rich hemolymph (Grbic et al. 1998;Grbic 2000; Wiegmann et al. 2011). The eggs ofsuch wasps are small, in the same size range as amammalian egg, and lack yolk. The early devel-opment of these small eggs differs dramaticallyfrom the canonical insect program. Notably, anew developmental module is inserted at thebeginning of development. The early mitoticcycles are greatly changed, and patterning eventsand gastrulation are delayed. During this period,the embryo grows dramatically reaching thesame size as the externally developing embryosthat began with vastly bigger eggs. The morphol-ogy of the gastrulating embryo differs from itsclose relatives; nonetheless, the endoparasiticwasps express classical insect-patterning genesin canonical patterns during this stage and gen-erate a phylotypic embryo resembling that ofits relatives. Although there is great diversityin the details, the overall features of the changesare similar in various lineages and are not unlikethe reorganization of early embryogenesis inmammals. Notably, it is clear that developmentis appropriately aligned at the phylotypic stage,and that a direct comparison of the cleavagecycles between even closely related specieswould give the impression that there was no re-lationship.





Many endoparasitic wasps lay only one eggin an individual host larva, but, in a number ofspecies, numerous progeny eventually emergeform this parasitized larvae. This is the resultof a remarkable adaptation in which the embryotwins repeatedly in a process called polyembry-ony (Grbic et al. 1998; Grbic 2000; Wiegmannet al. 2011). This process, which is particularlydramatic and well studied in the wasp Copido-soma floridanum, takes advantage of the growthstage that was inserted into the early embry-onic programs of the endoparasitic wasps. Thepregastrulation embryo of this species forms ahollow ball of growing and proliferating cells.The shell of cells subdivides to produce addi-tional balls as it grows. This process can producein the range of 2000 embryos before patterning,gastrulation, and differentiation begin. Thisprogram, which is of obvious advantage inproducing more wasps, is also an indication ofthe flexibility of the early developmental pro-gram once the growth constraint is removed.

Although the relatively close relationship ofendoparasitic wasps with other wasps providesa particularly clear illustration of the conse-quences of the release of the growth constraint,other parasites provide additional examples ofmodification of the ancestral early embryonicprogram. Flukes, or trematodes, comprise a sub-stantial class of parasitic flatworms with aboutfour times as many species as there are speciesof mammals. Usually, the primary or defini-tive host is a vertebrate. The parasites, often her-maphroditic, produce small encysted eggs thatare released from the infected primary host(Butcher et al. 2002). Typically, these eggs, ifingested by a mollusk, will develop into a prim-itive larva called a miracidium that takes innourishment from the host mollusk and growsand develops germinative tissue that generatesegg-like cells that enlarge and go through stagesrecognizable as cleavage, gastrulation, and es-tablishment of a phylotypic body plan (Pod-vyaznaya and Galaktionov 2014; Skala et al.2014). Notably, growth occurs throughoutthis embryogenesis. The miracidium nourishesmany embryos in a uterus-like sac. On release,the progeny, now called cercariae, can infectthe vertebrate host to begin a new cycle. Like

the egg of the wasp, the egg of the fluke managesto produce many embryos—polyembryony.However, instead of simple serial twining ofthe embryo, a complex zygotic program ofdevelopment in the molluskan host has beenadded to zygotically generate many egg-like cellsthat then initiate embryogenesis. This programsuggests that, once the growth constraint is re-leased, the earliest stage of development is evo-lutionarily plastic. In contrast to the dramaticchanges in early zygotic development, eventsclose to the phylotypic stage persist althoughin a context divorced from the usually close as-sociation with fertilization and initial divisionof the egg.

It is interesting to consider pluripotency inthe context of polyembryony. This lifestyle re-quires that embryonic cells grow and proliferatebut still retain pluripotency. The transcriptionfactors responsible for zygotic maintenance ofpluripotency in mammals are not conserved ininsects, which appear to use different transcrip-tion factors (e.g., zelda in Drosophila) to pro-mote early embryonic transcription. Perhapsthe endoparasitic wasps will have independentlyevolved a means of supporting pluripotencyduring growth and proliferation. For example,persistent expression of a Zelda-like transcrip-tion factor might provide a wasp embryo withthe equivalent of a pluripotency factor. Wemight expect that ES cell lines could be estab-lished from these organisms (see Jourdane andTheron 1980).

Overall, the rapid evolution of the early de-velopmental programs in endoparasitic waspssupports the interpretation that the growth con-straint played a major role in restraining suchdivergence. Despite occasional release of theconstraint, we expect that many aspects of theregulatory programs that control growth andproliferation were molded by an unchangingconstraint that influenced much of animal evo-lution.

SUMMARY

Consideration of oogenesis and early develop-ment across a wide swath of the animal kingdomsuggests that the issue of growth has been an

P.H. O’Farrell




enduring constraint influencing much of evolu-tion. It is suggested that this constraint emergesfrom a simple requirement that an externallydeposited egg be able to produce a feeding an-imal to have a workable life cycle. This demandwas met by production of large eggs. This thenproduced the widespread pattern of cell growthin phase 1 (oocyte production) and cell prolif-eration in phase 2 (early embryogenesis) of anoverall four-phase animal program of growthand proliferation. This solution stressed the ca-pacities of ordinary growth and division. Thegrowth of an oocyte and its division as an eggshows numerous accommodations to thesestresses. These accommodations are evident inrepresentatives of all animal phyla, arguing thatthe biology of these stages has been moldedby the pressures of this enduring constraint.Furthermore, embryonic development evolvedwith these imposed constraints, so the pro-grams of development are interwoven withthe programs for growth and proliferation. Fi-nally, I argue that mammalian developmentought to be viewed as a diverged branch ofthe mainline of animal evolution, which wasdramatically influenced by the release of thegrowth constraint that was originally forma-tive. Introduction of an early growth phase inmammalian embryogenesis required a novelfeature, persistent pluripotency of the prolifer-ating cells, which go on to ultimately supportthe deferred embryogenesis. Parasitic organ-isms, which also evade the growth constraint,show diverged early programs including smalleggs, early embryonic growth, and acquisitionof a proliferating pluripotent population ofcells. These examples of the release of the growthconstraint are an indication of the powerful in-fluence that this constraint has exerted on ani-mal evolution.

ACKNOWLEDGMENTS

I thank current and past members of the labo-ratory, especially Jeffrey Farrell, Jennifer Page,and Isaac Strong, as well as Danielle O’Farrellfor help in preparing this article. Our effortsare supported by Grants R37GM037193 andR01ES020725 to P.H.O.

REFERENCES

Bedford L. 1966. The electron microscopy and cytochemis-try of oogenesis and the cytochemistry of embryonic de-velopment of the prosobranch gastropod Bembicium na-num L. J Embryol Exp Morphol 15: 15–37.

Blobel G, Potter VR. 1967. Studies on free and membrane-bound ribosomes in rat liver. I: Distribution as related tototal cellular RNA. J Mol Biol 26: 279–292.

Bratic I, Hench J, Henriksson J, Antebi A, Burglin TR, Tri-funovic A. 2009. Mitochondrial DNA level, but not activereplicase, is essential for Caenorhabditis elegans develop-ment. Nucleic Acids Res 37: 1817–1828.

Brown DD, Dawid IB. 1968. Specific gene amplification inoocytes. Oocyte nuclei contain extrachromosomal repli-cas of the genes for ribosomal RNA. Science 160: 272–280.

Butcher AR, Brealey JK, Grove DI, Dymock RB. 2002. Bra-chylaima cribbi (Digenea: Brachylaimidae): Scanningelectron microscopical observations of the life-cycle stag-es. J Helminthol 76: 207–215.

Canon S, Fernandez-Tresguerres B, Manzanares M. 2011.Pluripotency and lineages in the mammalian blastocyst:An evolutionary view. Cell Cycle 10: 1731–1738.

Chase JW, Dawid IB. 1972. Biogenesis of mitochondria dur-ing Xenopus laevis development. Dev Biol 27: 504–518.

Coronado D, Godet M, Bourillot PY, Tapponnier Y, BernatA, Petit M, Afanassieff M, Markossian S, Malashicheva A,Iacone R, et al. 2013. A short G1 phase is an intrinsicdeterminant of naıve embryonic stem cell pluripotency.Stem Cell Res 10: 118–131.

Cox RT, Spradling AC. 2003. A Balbiani body and the fu-some mediate mitochondrial inheritance during Droso-phila oogenesis. Development 130: 1579–1590.

Del Pino EM, Humphries AA. 1978. Multiple nuclei duringearly oogenesis in Flectonotus pygmaeus and other mar-supial frogs. Biol Bull 154: 198–212.

Dobzhansky T. 1973. Nothing in biology makes sense exceptin the light of evolution. J Hered 68: 3–10.

Domazet-Loso T, Tautz D. 2010. A phylogenetically basedtranscriptome age index mirrors ontogenetic divergencepatterns. Nature 468: 815–818.

Eckelbarger KJ, Larson R. 1992. Ultrastructure of the ovaryand oogenesis in the jellyfish Linuche unguiculata andStomolophus meleagris, with a review of ovarian structurein the Scyphozoa. Marine Biol 114: 633–643.

Eckelbarger KJ, Young CM. 1997. Ultrastructure of the ovaryand oogenesis in the methane-seep mollusc Bathyneritanaticoidea (Gastropoda: Neritidae) from the Louisianaslope. Invertebr Biol 116: 299–312.

Edgar BA, O’Farrell PH. 1990. The three postblastoderm cellcycles of Drosophila embryogenesis are regulated in G2 bystring. Cell 62: 469–480.

Edgar BA, Schubiger G. 1986. Parameters controlling tran-scriptional activation during early Drosophila develop-ment. Cell 44: 871–877.

Edgar BA, Kiehle CP, Schubiger G. 1986. Cell cycle controlby the nucleo-cytoplasmic ratio in early Drosophila de-velopment. Cell 44: 365–372.

Edgar BA, Sprenger F, Duronio RJ, Leopold P, O’Farrell PH.1994. Distinct molecular mechanism regulate cell cycle





timing at successive stages of Drosophila embryogenesis.Genes Dev 8: 440–452.

Evans MJ, Kaufman MH. 1981. Establishment in culture ofpluripotential cells from mouse embryos. Nature 292:154–156.

Evans T, Rosenthal ET, Youngblom J, Distel D, Hunt T. 1983.Cyclin: A protein specified by maternal mRNA in seaurchin eggs that is destroyed at each cleavage division.Cell 33: 389–396.

Farrell JA, O’Farrell PH. 2013. Mechanism and regulation ofCdc25/twine protein destruction in embryonic cell-cycleremodeling. Curr Biol 23: 118–126.

Farrell JA, O’Farrell PH. 2014. From egg to gastrula: How thecell cycle is remodeled during the Drosophila mid-blas-tula transition. Annu Rev Genet 48: 269–294.

Follette PJ, O’Farrell PH. 1997. Connecting cell behavior topatterning: Lessons from the cell cycle. Cell 88: 309–314.

Follette PJ, Duronio RJ, O’Farrell PH. 1998. Fluctuations incyclin E levels are required for multiple rounds of endo-cycle S phase in Drosophila. Curr Biol 8: 235–238.

Fujii-Yamamoto H, Kim JM, Arai KI, Masai H. 2005. Cellcycle and developmental regulations of replication fac-tors in mouse embryonic stem cells. J Biol Chem 280:12976–12987.

Gall JG. 1968. Differential synthesis of the genes for ribo-somal RNA during amphibian oogenesis. Proc Natl AcadSci 60: 553–560.

Gall JG. 2012. Are lampbrush chromosomes unique to mei-otic cells? Chromosome Res 20: 905–909.

Gall JG, Wu Z, Murphy C, Gao H. 2004. Structure in theamphibian germinal vesicle. Exp Cell Res 296: 28–34.

Ganot P, Bouquet JM, Kallesøe T, Thompson EM. 2007. TheOikopleura coenocyst, a unique chordate germ cell per-mitting rapid, extensive modulation of oocyte produc-tion. Dev Biol 302: 591–600.

Ganot P, Moosmann-Schulmeister A, Thompson EM. 2008.Oocyte selection is concurrent with meiosis resumptionin the coenocystic oogenesis of Oikopleura. Dev Biol 324:266–276.

Gardner EJ, Nizami ZF, Talbot CC, Gall JG. 2012. Stableintronic sequence RNA (sisRNA), a new class of noncod-ing RNA from the oocyte nucleus of Xenopus tropicalis.Genes Dev 26: 2550–2559.

Gersh ES. 1968. Mutants at the bobbed locus in Drosophilamelanogaster: Relation to ribosomal RNA synthesis? Sci-ence 162: 1139.

Graham A, Richardson J. 2012. Developmental and evolu-tionary origins of the pharyngeal apparatus. Evodevo 3:24.

Grbic M. 2000. “Alien” wasps and evolution of develop-ment. Bioessays 22: 920–932.

Grbic M, Nagy LM, Strand MR. 1998. Development of poly-embryonic insects: A major departure from typical insectembryogenesis. Dev Genes Evol 208: 69–81.

Gruzova MN, Batalova FM. 1979. Nuclear structures in thetelotrophic ovarioles of nocturnal ground beetles (Tene-brionidae, Polyphaga). II: The oocyte nucleus of Blapslethifera and Gnaptor spinimanus. Light optical data. On-togenez 10: 323–331.

Gurdon JB, Elsdale TR, Fischberg M. 1958. Sexually matureindividuals of Xenopus laevis from the transplantation ofsingle somatic nuclei. Nature 182: 64–65.

Harvey EB. 1935. Cleavage without nuclei. Science 82: 277.

Hayashi S, Yamaguchi M. 1999. Kinase-independent activityof Cdc2/cyclin A prevents the S phase in the Drosophilacell cycle. Genes Cells 4: 111–122.

Howe JA, Howell M, Hunt T, Newport JW. 1995. Identifica-tion of a developmental timer regulating the stability ofembryonic cyclin A and a new somatic A-type cyclin atgastrulation. Genes Dev 9: 1164–1176.

Jourdane J, Theron A. 1980. Schistosoma mansoni: Cloningby microsurgical transplantation of sporocysts. Exp Para-sitol 50: 349–357.

Kalinka AT, Varga KM, Gerrard DT, Preibisch S, CorcoranDL, Jarrells J, Ohler U, Bergman CM, Tomancak P. 2010.Gene expression divergence recapitulates the develop-mental hourglass model. Nature 468: 811–814.

Kloc M, Jedrzejowska I, Tworzydlo W, Bilinski SM. 2014.Balbiani body, nuage and sponge bodies—The germplasm pathway players. Arthropod Struct Dev 43: 341–348.

Landstrom U, Løvtrup-Rein H, Løvtrup S. 1975. Control ofcell division and cell differentiation by deoxynucleotidesin the early embryo of Xenopus laevis. Cell Differ 4: 313–325.

Laskey RA, Honda BM, Mills AD, Finch JT. 1978. Nucleo-somes are assembled by an acidic protein which bindshistones and transfers them to DNA. Nature 275: 416–420.

Le Bin GC, Munoz-Descalzo S, Kurowski A, Leitch H, LouX, Mansfield W, Etienne-Dumeau C, Grabole N, MulasC, Niwa H, et al. 2014. Oct4 is required for lineage prim-ing in the developing inner cell mass of the mouse blas-tocyst. Development 141: 1001–1010.

Lee MT, Bonneau AR, Takacs CM, Bazzini AA, DiVito KR,Fleming ES, Giraldez AJ. 2013. Nanog, Pou5f1 and SoxB1activate zygotic gene expression during the maternal-to-zygotic transition. Nature 503: 360–364.

Leichsenring M, Maes J, Mossner R, Driever W, Onicht-chouk D. 2013. Pou5f1 transcription factor controls zy-gotic gene activation in vertebrates. Science 341: 1005–1009.

Li XY, Harrison MM, Villalta JE, Kaplan T, Eisen MB. 2014.Establishment of regions of genomic activity during theDrosophila maternal to zygotic transition. eLife doi: 10.7554/eLife.03737.

Lin CJ, Koh FM, Wong P, Conti M, Ramalho-Santos M.2014. Hira-mediated H3.3 incorporation is required forDNA replication and ribosomal RNA transcription in themouse zygote. Dev Cell 30: 268–279.

Loppin B, Bonnefoy E, Anselme C, Laurencon A, Karr TL,Couble P. 2005. The histone H3.3 chaperone HIRA isessential for chromatin assembly in the male pronucleus.Nature 437: 1386–1390.

Mac Auley A, Werb Z, Mirkes PE. 1993. Characterization ofthe unusually rapid cell cycles during rat gastrulation.Development 117: 873–883.

Macgregor HC, Del Pino EM. 1982. Ribosomal gene ampli-fication in multinucleate oocytes of the egg brooding

P.H. O’Farrell




hylid frog Flectonotus pygmaeus. Chromosoma 85: 475–488.

Marandel L, Labbe C, Bobe J, Le Bail PY. 2012. nanog 50-upstream sequence, DNA methylation, and expression ingametes and early embryo reveal striking differences be-tween teleosts and mammals. Gene 492: 130–137.

Marlow FL, Mullins MC. 2008. Bucky ball functions in Bal-biani body assembly and animal–vegetal polarity in theoocyte and follicle cell layer in zebrafish. Dev Biol 321:40–50.

Martin GR. 1981. Isolation of a pluripotent cell line fromearly mouse embryos cultured in medium conditionedby teratocarcinoma stem cells. Proc Natl Acad Sci 78:7634–7638.

Newport J, Dasso M. 1989. On the coupling between DNAreplication and mitosis. J Cell Sci 12: 149–160.

Newport J, Kirschner M. 1982a. A major developmentaltransition in early Xenopus embryos. I: Characterizationand timing of cellular changes at the midblastula stage.Cell 30: 675–686.

Newport J, Kirschner M. 1982b. A major developmentaltransition in early Xenopus embryos. II: Control of theonset of transcription. Cell 30: 687–696.

O’Farrell PH. 2004. How metazoans reach their full size: Thenatural history of bigness. In Cell growth: Control of cellsize (ed. Hall M, et al.). Cold Spring Harbor LaboratoryPress, Cold Spring Harbor, NY.

O’Farrell PH. 2011. Quiescence: Early evolutionary originsand universality do not imply uniformity. Phil Trans RSoc B 366: 3498–3507.

O’Farrell PH, Stumpff J, Su TT. 2004. Embryonic cleavagecycles: How is a mouse like a fly? Curr Biol 14: R35–R45.

Orsi GA, Algazeery A, Meyer RE, Capri M, Sapey-TriompheLM, Horard B, Gruffat H, Couble P, Aıt-Ahmed O, Lop-pin B. 2013. Drosophila Yemanuclein and HIRA cooper-ate for de novo assembly of H3.3-containing nucleo-somes in the male pronucleus. PLoS Genet 9: e1003285.

Patel NH, Martin-Blanco E, Coleman KG, Poole SJ, EllisMC, Kornberg TB, Goodman CS. 1989. Expression ofengrailed proteins in arthropods, annelids, and chor-dates. Cell 58: 955–968.

Pepling ME, de Cuevas M, Spradling AC. 1999. Germlinecysts: A conserved phase of germ cell development?Trends Cell Biol 9: 257–262.

Philpott A, Leno GH, Laskey RA. 1991. Sperm decondensa-tion in Xenopus egg cytoplasm is mediated by nucleoplas-min. Cell 65: 569–578.

Philpott A, Krude T, Laskey RA. 2000. Nuclear chaperones.Semin Cell Dev Biol 11: 7–14.

Podvyaznaya IM, Galaktionov KV. 2014. Trematode repro-duction in the molluscan host: An ultrastructural study ofthe germinal mass in the rediae of Himasthla elongata(Mehlis, 1831) (Digenea: Echinostomatidae). ParasitolRes 113: 1215–1224.

Pritchard DK, Schubiger G. 1996. Activation of transcrip-tion in Drosophila embryos is a gradual process mediatedby the nucleocytoplasmic ratio. Genes Dev 10: 1131–1142.

Prud’homme B, Gompel N. 2010. Evolutionary biology:Genomic hourglass. Nature 468: 768–769.

Prud’homme B, de Rosa R, Arendt D, Julien JF, Pajaziti R,Dorresteijn AWC, Adoutte A, Wittbrodt J, BalavoineG. 2003. Arthropod-like expression patterns of en-grailed and wingless in the annelid Platynereis dumeriliisuggest a role in segment formation. Curr Biol 13: 1876–1881.

Raff RA. 1996. The shape of life. University of ChicagoPress, Chicago.

Rahn H, Paganelli CV, Ar A. 1974. The avian egg: Air-cell gastension, metabolism and incubation time. Respir Physiol22: 297–309.

Sanchez-Sanchez AV, Camp E, Mullor JL. 2011. Fishingpluripotency mechanisms in vivo. Int J Biol Sci 7: 410–417.

Savatier P, Lapillonne H, van Grunsven LA, Rudkin BB,Samarut J. 1996. Withdrawal of differentiation inhibitoryactivity/leukemia inhibitory factor up-regulates D-typecyclins and cyclin-dependent kinase inhibitors in mouseembryonic stem cells. Oncogene 12: 309–322.

Shen MM. 2007. Nodal signaling: Developmental roles andregulation. Development 134: 1023–1034.

Shermoen AW, O’Farrell PH. 1991. Progression of the cellcycle through mitosis leads to abortion of nascent tran-scripts. Cell 67: 303–310.

Shermoen AW, McCleland ML, O’Farrell PH. 2010. Devel-opmental control of late replication and S phase length.Curr Biol 20: 2067–2077.

Skala V, Bulantova J, Walker AJ, Horak P. 2014. Insights intothe development of Notocotylus attenuatus (Digenea: No-tocotylidae) in Lymnaea stagnalis: From mother sporo-cyst to cercariae. Parasitol Int 63: 94–99.

Smiley S. 1990. A review of echinoderm oogenesis. J ElectronMicrosc Tech 16: 93–114.

Snow MHL. 1977. Gastrulation in the mouse: Growth andregionalization of the epiblast. J Embryol Exp Morphol 42:293–303.

Snow MH. 1981. Growth and its control in early mamma-lian development. Br Med Bull 37: 221–226.

Sun LT, Yamaguchi S, Hirano K, Ichisaka T, Kuroda T, Tada T.2014. Nanog co-regulated by Nodal/Smad2 and Oct4 isrequired for pluripotency in developing mouse epiblast.Dev Biol 392: 182–192.

Takahashi K, Yamanaka S. 2006. Induction of pluripotentstem cells from mouse embryonic and adult fibroblastcultures by defined factors. Cell 126: 663–676.

Tsang WY, Lemire BD. 2002. Mitochondrial genome con-tent is regulated during nematode development. BiochemBiophys Res Commun 291: 8–16.

Tworzydlo W, Kloc M, Bilinski SM. 2009. The Balbiani bodyin the female germline cells of an earwig, Opisthocosmiasilvestris. Zool Sci 26: 754–757.

van der Laan S, Tsanov N, Crozet C, Maiorano D. 2013. HighDub3 expression in mouse ESCs couples the G1/S check-point to pluripotency. Mol Cell 52: 366–379.

Webb AC, Smith LD. 1977. Accumulation of mitochondrialDNA during oogenesis in Xenopus laevis. Dev Biol 56:219–225.

White J, Stead E, Faast R, Conn S, Cartwright P, Dalton S.2005. Developmental activation of the Rb-E2F pathwayand establishment of cell cycle-regulated cyclin-depen-





dent kinase activity during embryonic stem cell differen-tiation. Mol Biol Cell 16: 2018–2027.

Wiegmann BM, Trautwein MD, Winkler IS, Barr NB, KimJW, Lambkin C, Bertone MA, Cassel BK, BaylessKM, Heimberg AM, et al. 2011. Episodic radiationsin the fly tree of life. Proc Natl Acad Sci 108:5690– 5695.

Wieschaus E. 1996. Embryonic transcription and the con-trol of developmental pathways. Genetics 142: 5–10.

Wolf SF, Schlessinger D. 1977. Nuclear metabolism ofribosomal RNA in growing, methionine-limited, andethionine-treated HeLa cells. Biochemistry 16: 2783–2791.

Woodland H. 1982. The translational control phase of earlydevelopment. Biosci Rep 2: 471–491.

Woodland HR, Pestell RQ. 1972. Determination of the nu-cleoside triphosphate contents of eggs and oocytes ofXenopus laevis. Biochem J 127: 597–605.

Yang VS, Carter SA, Hyland SJ, Tachibana-Konwalski K,Laskey RA, Gonzalez MA. 2011. Geminin escapes degra-dation in G1 of mouse pluripotent cells and mediates theexpression of Oct4, Sox2, and Nanog. Curr Biol 21: 692–699.

Yuan K, Shermoen AW, O’Farrell PH. 2014. IlluminatingDNA replication during Drosophila development usingTALE-lights. Curr Biol 24: R144–R145.

Zhao ZK, Li W, Wang MY, Zhou L, Wang JL, Wang YF. 2011.The role of HIRA and maternal histones in sperm nucleusdecondensation in the gibel carp and color crucian carp.Mol Reprod Dev 78: 139–147.

P.H. O’Farrell




published online August 7, 2015Cold Spring Harb Perspect Biol Patrick H. O'Farrell Growing an Embryo from a Single Cell: A Hurdle in Animal Life

Subject Collection Size Control in Biology: From Organelles to Organisms

Cell-Size ControlAmanda A. Amodeo and Jan M. Skotheim

Mechanical Forces and Growth in Animal TissuesLoïc LeGoff and Thomas Lecuit

Ancestral Condition?Indeterminate Growth: Could It Represent the

H. WakeIswar K. Hariharan, David B. Wake and Marvalee

AmphibiansBiological Scaling Problems and Solutions in

Daniel L. Levy and Rebecca Heald

The Systemic Control of GrowthLaura Boulan, Marco Milán and Pierre Léopold

Intracellular Scaling MechanismsSimone Reber and Nathan W. Goehring

DiversityGenome Biology and the Evolution of Cell-Size

Rachel Lockridge Muellerin Animal LifeGrowing an Embryo from a Single Cell: A Hurdle

Patrick H. O'Farrell

EmbryosXenopusSize Scaling of Microtubule Assemblies in Early

Nguyen, et al.Timothy J. Mitchison, Keisuke Ishihara, Phuong

Organ-Size Regulation in MammalsAlfredo I. Penzo-Méndez and Ben Z. Stanger

Morphology in AmphibiansThe Influence of Genome and Cell Size on Brain

Gerhard Roth and Wolfgang WalkowiakFlowers

Lessons from Leaves and−−Size Control in Plants

Hjördis Czesnick and Michael Lenhard

Controland AMP-Activated Protein Kinase in Cell Growth The Opposing Actions of Target of Rapamycin

N. HallSravanth K. Hindupur, Asier González and Michael

Human Cell TypesNuclear DNA Content Varies with Cell Size across

DamianiJames F. Gillooly, Andrew Hein and Rachel

Small but Mighty: Cell Size and BacteriaPetra Anne Levin and Esther R. Angert

Subcellular SizeWallace F. Marshall

http://cshperspectives.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2015 Cold Spring Harbor Laboratory Press; all rights reserved


http://cshperspectives.cshlp.org/cgi/collection/

http://cshperspectives.cshlp.org/cgi/collection/


growing an embryo from a single cell: a hurdle in animal...

Documents