NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Marlow FL. Maternal Control of Development in Vertebrates: My Mother Made Me Do It! San Rafael (CA): Morgan & Claypool Life Sciences; 2010.

Cover of Maternal Control of Development in Vertebrates

Maternal Control of Development in Vertebrates: My Mother Made Me Do It!

Show details


“My mother made me do it!” Most of us have uttered this phrase at least once. It is commonly used in the years of our life when we are transitioning from dependent child to independence, have our own ideas, and are trying to take control of our own decisions for better or for worse. Use of the phrase was generally induced anytime we felt that our will was suppressed by our mother’s will. Teenagers know all about maternal control and how it affects their phenotype, for example, how their social development is impaired when they cannot wear skinny jeans, low cut tops, or open Facebook accounts. Teenagers also have their own mechanisms to modulate the output of the maternal program. For example, the outfit that technically was not worn to school, but was instead stuffed into a backpack—the picture day photos do not lie. All psychology aside, none of us complain about or probably even think about the time during our development when our mother truly was in control, the period between production and fertilization of our egg and activation of our own zygotic genome. During this time, we dutifully followed our mothers’ instructions; never mind that we did not have a brain when we were fertilized eggs or cleaving embryos. This review focuses on maternal regulation of early vertebrate development.

Multicellular organisms develop from a single fertilized cell. This cell is endowed with the potential to generate more cells that will interact with one another through direct contact and communicate through highly conserved signaling pathways to build an embryo with defined axes and organ systems. The embryo will continue to develop and grow into a mature adult capable of producing the cells necessary to form the next generation. In animals, this cellular interplay involves the formation of the gametes within the gonads of developing animals. Males of a species produce sperm, while females produce the larger of the gametes, the oocytes, and hermaphrodites produce both sperm and oocytes. In each sex, sperm or egg production depends on interactions and signaling between the somatic and germ line derived cells that comprise the gonad. Both cell types are ultimately crucial for gamete production, and thus for restoring the diploid DNA content through fertilization and to ensure optimal development of the resulting embryo. This review will focus on the contribution of the maternal gamete, the oocyte, which develops as the egg; when fertilized, this cell has the capacity to form all the cells of the embryo.

In addition to supplying half the DNA, eggs are full of nutrients, which in animals with large yolky eggs support the needs of the embryo until it can acquire food on its own. Eggs of all animals also contain mRNAs and proteins that are supplied to or deposited in the egg as it develops during oogenesis. These maternal gene products regulate meiosis, oocyte development, and early development of the embryo including fertilization, transitions between meiotic and mitotic cell cycles, and the switch from utilization of mRNAs and proteins provided by the mother to the embryo’s own gene products during zygotic genome activation. The signals and regulation of meiosis will not be discussed in detail. However, many stages and aspects of oocyte development and egg production are conserved among vertebrates and will be touched upon. Here, each chapter emphasizes a maternally regulated process and the essential maternal genes required for that aspect of vertebrate development.

Maternal-Effect Genes

Maternal genes are those genes whose products, RNA or protein, are produced or deposited in the oocyte or are present in the fertilized egg or embryo before expression of zygotic genes is initiated. Genes whose RNA and protein products are only produced after zygotic genome activation, at or after the maternal to zygotic transition stage, are zygotic genes. Microarray expression studies have revealed numerous genes that are maternally expressed in vertebrates (Dworkin and Dworkin-Rastl, 1990; Schultz and Heyner, 1992; Wassarman and Kinloch, 1992; Temeles et al., 1994; Zimmermann and Schultz, 1994; Nothias et al., 1995; Wiekowski et al., 1997; Pelegri, 2003; Mager et al., 2006; White and Heasman, 2008). While this expression information is suggestive of function and provides comparative insight into which genes are exclusively maternal or zygotic, only functional investigation of maternally expressed genes can tell us which genes have essential maternal functions during development. It is worth noting that there are also examples of genes which are not maternally expressed but whose zygotic transcripts begin to be detected at stages prior to global activation of the zygotic genome, some of which regulate clearance of maternal mRNAs (Yang et al., 2002; Leung et al., 2003; Zhang et al., 2003; Lund et al., 2009). The mechanisms mediating the selective activation of genes prior to zygotic genome activation and in the context of global repression are intriguing and not well understood.

Investigating Maternal Function

An embryo exclusively relies on maternal gene products, RNAs, and proteins for its early development until activation of its own genome. The precise developmental time period and developmental processes under maternal control when the embryo is largely transcriptionally silent vary among organisms. In some animals, such as mice, humans, and nematodes (Caenorhabditis elegans), only the first or first couple of cleavage cycles are accomplished before transcription of the embryonic genome is activated. Other organisms, such as Drosophila, Xenopus, and zebrafish, rely on maternal RNAs and proteins for a more prolonged developmental period that includes several additional rounds of cleavage cycles. This dependence on maternal products persists to include regulation of early embryonic patterning, morphogenesis, and even extends to developmental processes occurring well after most maternal gene products are expected to have been degraded (e.g., mitotic divisions in mouse, late gastrulation, segmentation, and pattern formation in zebrafish and Xenopus). Specific examples of persisting maternal function will be discussed throughout this review and in the chapters on persisting maternal function.

What is a maternal effect? Recessive mutations disrupting zygotic gene function produce visible phenotypes only when the embryo is of homozygous mutant genotype and, thus, has two copies of the mutated or abnormal gene. Embryos with only one copy of the mutant allele are heterozygous and so appear normal. If a zygotically required gene is also maternally expressed, the maternal gene products can weaken the developmental consequences even when zygotic gene function is absent in the genotypically homozygous mutant progeny. In such cases, the embryos will have a milder defect than would otherwise be predicted based on gene expression pattern or abundance because the maternal product perdures and compensates for impaired zygotic gene function. Several examples of maternal compensation will be discussed further in the later chapters.

In contrast to zygotic recessive mutations, where visible phenotypes are observed only in homozygous individuals (Figure 1), homozygous mutant individuals for genes with strict maternal functions (i.e., no zygotic requirement) appear morphologically normal. The normal appearance of the embryos is due to the normal gene function contributed by their heterozygous mother (Figure 1). For genes with strict maternal function, mutant males are also of normal phenotype and, in the majority of cases, are fertile carriers of the defective gene. Mutant females are also morphologically normal until they reach reproductive maturity. In this review, a female is not sterile if she is able to produce mature eggs that can be fertilized, although the female is technically sterile in that her progeny are not competent to undergo normal development. All of the eggs or offspring produced by these mutant females will show developmental defects, even if they are genotypically heterozygous (Figure 1). For these progeny, loss of the maternally contributed product or failure of the mutant mother to provide normal maternal gene function is thus in effect dominant; these mutations are called maternal-effect lethals.

FIGURE 1. Genetic screens to uncover maternal-effect genes and maternally regulated processes.


Genetic screens to uncover maternal-effect genes and maternally regulated processes. In zebrafish, males are mutagenized (G0). Mutagenized males are crossed with females of the same genetic background to generate F1 carries of newly introduced mutations. (more...)

When we consider that maternal gene products are produced during oogenesis and that development of an embryo requires the product of meiosis, a developmentally competent egg, it is reasonable to consider genes essential for oocyte development among genes with maternal function and thus to include oogenesis as a maternal-effect process. However, the focus of this review will be largely limited to genes, or mutant alleles of those genes that are not essential for^egg^production, but instead impact patterning, fertilization, activation of the egg, and later aspects of embryonic development. While mutations compromising uterine competence in mammals, such as p53 and its target leukemia inhibitory factor, play key roles in maternal reproduction, these molecules will not be discussed in detail in this review (Hu et al., 2008; Hu et al., 2007; Kang et al., 2009). Reproductive defects will be included among the maternal-effect genes if they are localized to the germline and do not preclude formation of a fertilization competent egg.

Targeting Maternal-Effect Genes

The source of maternal RNAs and proteins include molecules produced by the developing oocyte as well as somatic cells, which produce products that are later imported into the oocyte. In Drosophila, C. elegans, Xenopus, fish, and mammals proteins or their precursors are imported via the maternal bloodstream or from the somatic cells of the follicle. These gene products together with the RNAs and proteins produced by the oocyte constitute the maternal pool. The abundance of maternal proteins and transcripts in the developing oocytes within the vertebrate ovary makes maternal products difficult targets for interference because manual injection of interfering forms of the gene products, either plasmid DNA, messenger mRNA, or antibodies against the protein of interest are difficult to deliver or cannot be delivered in time to effectively block gene function before it is required.

Recently, several methods have made it possible to target or interfere with maternal gene function in model organisms. These approaches and some advantages and limitations of each are discussed briefly and are summarized in Table 1. Due to variability in the effectiveness of each approach in distinct model systems and stage-specific limitations due to oocyte size, number, or accessibility, it remains necessary to develop and utilize multiple approaches tailored to a specific model system, developmental process, or stage to study the maternal function of a candidate gene.

TABLE 1. Methods used to interfere with maternal gene function in model systems.


Methods used to interfere with maternal gene function in model systems.


In contrast to the limitations of transient depletion, mutants, embryos with genetic lesions caused by chemical mutagens, radiation, retroviral insertion, or zinc finger nucleases disrupting genes required maternally for development of oocytes or early embryos allow perturbation of gene function at the earliest stage when it is required without causing off-target effects. Such maternal-effect mutants have been discovered in large-scale mutagenesis screens in model organisms including Drosophila melanogaster (flies), C. elegans (worms), Danio rerio (fish) and made by homologous recombination to disrupt candidate genes in Mus musculus (mouse) (Table 1) reviewed by St Johnston and Nusslein-Volhard (1992), Kemphues and Strome (1997), Pelegri (2003), Pelegri and Mullins (2004), Abrams and Mullins (2009), and Lindeman and Pelegri (2009). Though maternal-effect screens have only recently been carried out in vertebrates, such mutants have already identified novel maternal-effect functions for genes with essential maternal roles during early embryonic development. The maternal-effect screens in zebrafish yielded the largest collection of maternal-effect mutants in vertebrates; nevertheless, only a fraction of the total expected were identified, based on the number of genomes screened and estimates from previous large-scale zygotic screens (Mullins et al., 1994; Solnica-Krezel et al., 1994; Driever et al., 1996). Specifically, after screening 600 genomes, the vast majority of zygotic mutations were represented by single alleles, and less than one-half of the expected genes were identified. The zebrafish maternal-effect screens combined to date approach these numbers; however, these screens select against maternal genes that also have essential zygotic functions (Dosch et al., 2004; Luschnig et al., 2004; Pelegri et al., 2004).

Germline Replacement: Bypassing Essential Zygotic Functions

Germline replacement has been a successful approach to investigate the maternal function of genes with essential zygotic requirements during zebrafish development (Ciruna et al., 2002). Briefly, this approach involves elimination of the host germline using a morpholino (discussed further in the next section) to prevent production of a gene product essential for survival of the germline stem cells (Ciruna et al., 2002). The germline of the host embryo is replaced by transplantation of labeled germline stem cells of a mutant donor. The resulting fish have wild-type somatic composition and mutant germline tissues; thus, allowing examination of the maternal function of the mutated gene. This approach is very effective, but is also labor intensive. Therefore, alternative approaches have been pursued to further access the essential maternal genes and to improve our understanding on the molecular and cellular basis of maternally regulated aspects of vertebrate development.

Morpholinos and Oligo RNaseH

Advanced or maturation competent stage oocytes can be harvested from the ovaries of some vertebrate females, including mice and frogs, and injected with an interfering molecule, for example, a dominant negative form of a protein, morpholinos (antisense oligonucleotides designed to either block splicing of messages or translation of proteins from mRNAs) (Coonrod et al., 2001; Kanzler et al., 2003), or by injecting stabilized versions of antisense DNA oligonucleotides to induce RNaseH-mediated depletion of maternal mRNAs (Table 1) (Coonrod et al., 2001; Mir and Heasman, 2008). Once injected, the oocytes are induced to undergo maturation. This is accomplished either by returning the oocytes to a surrogate mother or by inducing maturation by supplying hormones to cells in culture. Finally, the manipulated oocytes are fertilized, via natural or in vitro approaches, and the resulting embryos are examined for developmental abnormalities. These methods work well in late-stage oocytes of Xenopus and mouse model systems (for more detailed methodology, the reader is referred to Colledge et al., 1994 and Mir and Heasman, 2008). Late-stage vertebrate oocytes are large and relatively easy to manipulate compared with primary oocytes. Similar approaches to conduct an interference study in which early-stage oocytes are manipulated, cultured to maturation-competent stages, fertilized to examine the consequences to early development of the embryo have not been feasible thus far. Despite the limited developmental potential of manipulated oocytes, these approaches are a useful means to obtain a snapshot of how a gene contributes to a particular stage of oocyte development. For example, mRNA, morpholino, antibodies, or toxin can be injected into oocytes to examine the effect on mRNA localization in early- or late-stage oocytes and the ability of the oocyte to undergo maturation or to eliminate a gene product in oocytes to examine its role in the early development of the embryo.


The RNA interference pathway, which regulates gene activity and acts to protect cells against pathogens, has been successfully co-opted by researchers as a strategy for depleting gene function by interfering with maternal messages in oocytes (Svoboda et al., 2000; Svoboda et al., 2001). Double-stranded RNA, dsRNA, targeting a gene of interest, is introduced to the animal by feeding, by soaking, or introducing hairpin-producing transgenes or viruses by direct microinjection or transfection. The consequence is degradation or silencing of the target gene product, and in some organisms, such as the worm, this silencing is transmitted through the germline (Fire et al., 1998). RNAi is not very effective in translationally quiescent cells; however, when transcribed messages become translationally active during oocyte maturation in late stages of oogenesis, RNAi is an effective means to interfere with gene function (Svoboda et al., 2000; Svoboda et al., 2001; Kennerdell et al., 2002). A clear advantage of the RNAi approach is that the animal and the gonad are not physically manipulated or removed from the animal during the depletion, thus limiting unintended and undesired nonspecific perturbations due to physical manipulation.

RNAi hairpin technology has recently been reported to be effective in zebrafish (Dong et al., 2009). The strategy takes advantage of the efficient production of short hairpin RNAs by the miR30 transcribed region when expressed from a ubiquitous or tissue-specific promoter (Dong et al., 2009). The endogenous hairpin is replaced with a hairpin targeted to a gene of interest, which when processed will produce a heritable transgene capable of producing interfering RNAs. Although this shRNAi strategy has not yet been applied to study a maternal-effect function, when combined with tissue-specific promoters, this knockdown strategy should improve access to candidate genes with both essential zygotic and maternal functions.

Together, these molecular and genetic approaches provide access to maternally regulated processes that are conserved yet poorly understood, including features of oocyte and early embryonic development.

Copyright © 2010 by Morgan & Claypool Life Sciences.
Bookshelf ID: NBK53192


  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...