Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2007 Jan 30;104(6):1745–1750. doi: 10.1073/pnas.0610108104

Embryo stability and vulnerability in an always changing world

Amro Hamdoun 1,*, David Epel 1,*
PMCID: PMC1794293  PMID: 17264211

Abstract

Contrary to the view that embryos and larvae are the most fragile stages of life, development is stable under real-world conditions. Early cleavage embryos are prepared for environmental vagaries by having high levels of cellular defenses already present in the egg before fertilization. Later in development, adaptive responses to the environment either buffer stress or produce alternative developmental phenotypes. These buffers, defenses, and alternative pathways set physiological limits for development under expected conditions; teratology occurs when embryos encounter unexpected environmental changes and when stress exceeds these limits. Of concern is that rapid anthropogenic changes to the environment are beyond the range of these protective mechanisms.


A widely held view, as articulated by J. G. Wilson in 1973, has been that “living things during early developmental stages are more sensitive than at any other time in their life cycle to adverse influences in the environment” (1). Early 20th century studies in mammals demonstrated how the environment, rather than heredity alone, could cause birth defects. Later, the perception of embryo vulnerability was reinforced by the real-life example of thalidomide, a drug first administered to women in the 1950s and subsequently found to be responsible for severe developmental malformations in thousands of children. More recently, studies have extended our view of teratology to include adverse effects of exposure to stresses in the womb that are manifested much later in life.

Yet equally striking are examples of how real-world development of animal embryos works normally, despite formidable environmental challenges (2). We review evidence indicating that the keys for developmental success are cellular mechanisms that provide robustness and buffer embryos from the environment and regulatory pathways that alter the developmental path in response to the conditions encountered. These mechanisms provide potent, although not impregnable, defenses against common stressors in development including changes in temperature, hypoxia, pathogens, UV radiation, free radicals, and toxicants. These defenses and pathways have evolved to permit optimal development in the environment likely to be encountered, with limits adapted to the historical environment of the embryos.

We suggest a view of embryo vulnerability that takes into account our emerging appreciation of the potent protections present in embryos. We posit that, although embryos are well buffered for expected environment(s), rapid anthropogenic changes can overwhelm this intrinsic robustness. Of special concern are man-made chemicals that evade developmental defenses or misdirect developmental decisions. Finally, we provide examples of how an understanding of the mechanisms that underlie developmental stability can be used to predict environmental changes that are problematic for embryos.

Protection Strategies in Early Development: A General Strategy for Protection

The survival problem for the early embryo is that these few-celled rudiments do not possess sophisticated adult defense systems such as the immune system or the nervous system. One approach, especially developed in the mammals, is parental protection of the embryo. Nevertheless, as illustrated by the thalidomide example, even structures such as the placenta do not provide complete protection.

Most organisms, however, do not rely on parental protection. Instead, they release eggs or embryos into their surroundings to develop as “orphans” in direct contact with the environment. One adaptation in these embryos is to accelerate development and speed progression through these undifferentiated early stages, through rapid cell divisions (3).

The second and major adaptation of these orphan embryos is to package high levels of cellular defenses into the egg prior to release into the environment and before exposure to stress (Fig. 1). As we will describe below, the specific defenses in embryos of different species are the product of selection for a “be prepared” strategy. This selection is in response to the past history for each species and results in adaptations against the stresses its embryos might encounter during development. Some of these strategies are specifically for protection and are independent of developmental programs; other protections are also integral parts of developmental programs.

Fig. 1.

Fig. 1.

In early development (gray background) of embryos with rapid early cleavages (e.g., most orphan embryos) the strategy is to have high levels of innate defenses already present before encountering environmental stress. Later in development, the embryo can respond to the environment by inducing defenses, altering developmental path, or temporarily stopping development. Under severe conditions the alternate paths can represent a “lottery approach,” producing phenotypes that may or may not be adaptive.

Protective Strategies Independent of Developmental Programs

Protection from UV Radiation.

Many embryos of aquatic organisms develop at or near the air–water interface and, hence, are exposed to high levels of UV radiation (UV) which can directly damage DNA as well as produce harmful free radicals. A major adaptation is to incorporate UV-A- and UV-B-absorbing compounds into the eggs (reviewed in ref. 4). Mycosporine amino acids are one example of these agents; evidence for their protective role is that sea urchin embryos with low levels of these diet-derived sunscreens are more susceptible to UV damage than embryos with high levels of these compounds (5).

Although the sunscreens clearly provide some protection, other adaptations are used in high-UV conditions. The embryos of an Antarctic sea urchin are irreversibly damaged if placed near the surface under polar summer sunlight, and the major damage is from the UV-A part of the spectrum. However, these embryos are normally nonbuoyant, so the additional adaptation here is production of eggs that sink to the bottom and thus remain out of the range of damaging UV (6).

Recent research suggests that global climate change can indirectly increase penetration of UV into shallow lakes and streams, with potentially harmful consequences for embryos. For example, toad embryos, Bufo boreas, develop in the high mountain lakes of the Cascade mountains, and their susceptibility to infection by the fungus Saprolignia ferix is increased in drought years during which lakes are shallower, and UV exposure is greater (7). The frequency and severity of these drought years is itself a consequence of the increased sea surface temperatures in the Pacific, concomitant with rapid global climate changes seen since the mid 1970s.

Pathogen Defenses.

A second example of environment-specific defenses is innate protection against bacterial and fungal pathogens. One strategy is physical barriers to pathogens, such as from egg chorions and shells. Another strategy centers on the accumulation, during oogenesis, of antibacterial substances in the egg and extraembryonic tissues. A classic example is the antibacterial muramidase enzymes, first described in bird eggs (8); these enzymes attack carbohydrates on the surface of pathogenic bacteria. More recent examples include proteins such as the defensins, which, along with other potent antibacterial peptides, are present in the tissues surrounding early stages of the tobacco hornworm embryo (9). Microbial defenses are present in fish embryos, where fertilization results in release of antibacterial and antifungal enzymes from granules in the egg cortex (10).

Fungal infections are a major problem for marine organisms, and some embryos deter fungi with the products of the symbiotic bacteria they harbor. Crustacean and squid embryos harbor symbiotic bacteria on the egg surface or in embryo sheaths. In the shrimp Palaeomon macrodactylus, symbiotic Alteromonas produce indolinedione, which protects the embryo from fungal infections (11). In squid, bacteria live as symbionts in the adult accessory nidamental gland, and the bacteria are secreted onto the egg jelly and chorion at spawning (12).

Physical Protection.

Eggs and embryos must maintain physical integrity before fertilization and during development. The primary protection comes from extraembryonic structures in the egg, such as chorions, jellies, and shells, which can account for a significant proportion of the egg's mass. In the sea urchin, for example, ≈10% of the total protein of the egg is present in the cortical granules that exocytose at fertilization and help form the protective fertilization envelope; if this envelope is removed, the embryo is more susceptible to physical damage (13). Similarly, removal of jellies around sea urchin eggs renders them susceptible to damage by the shear stresses encountered during spawning and from waves in their marine environment (14).

Another protection comes from a mechanism to repair physical disruption or punctures to the plasma membrane. Embryologists make use of this whenever they microinject eggs or embryos; the cells survive, and the embryos develop normally. The protective mechanism is calcium-activated membrane sealing (15). When the membrane is broken, the influx of extracellular calcium ions triggers fusion of intracellular vesicles at the site of injury (15, 16). The added vesicle membranes plug and fuse with the injured membrane so that embryo integrity is maintained. As with jellies and chorions, eggs and early embryos possess large amounts of these intracellular membrane vesicles, suggesting that physical damage can be a major problem for embryos.

Protective Strategies Integral to Developmental Programs

Defenses Against Xenobiotic Chemicals.

Exposure to toxic chemicals can be an expected problem in development. They occur naturally, and embryos can be exposed directly, as with exposure of orphan embryos to phytotoxins and microbial byproducts, or indirectly, as seen with passage of dietary toxins through the bloodstream and placenta in mammals.

In the early developmental stages, toxicant transformation by the cytochrome P450 (CYPS) has not been observed, and, indeed, xenobiotic transforming CYPS do not appear until differentiation of tissues. One possibility for their absence in early stages may be that their activity can bioactivate otherwise benign metabolic products into mutagens (17).

Rather, the primary defense is to keep the toxicants out of the embryo by using membrane transporters, such as the multidrug-resistance proteins (MRPs). These efflux transporters, which might be thought of as first lines of defense, are ABC proteins that include the ABCB/PGP (18), ABCC/MRP (19), and ABCG/MXR (20) transporters.

An example of the use of efflux transporters for defense is seen in the echiuroid worm Urechis caupo, which develop in mudflats containing metabolic byproducts of bacteria and plant-degradation products. These embryos have levels of PGP efflux activity even higher than those seen in a drug-resistant, transporter-overexpressing cancer cell (21). The PGP transporters are also essential for protection of mammalian embryos. Mice and rats have high levels of PGP transporters in the placenta (22) and in the embryo (23). Embryos of transgenic mice lacking PGP accumulate 5-fold-higher levels of toxic compounds administered to the mothers than their normal, PGP-expressing counterparts. A similar increase in fetal toxicant accumulation is seen in normal mice that are fed drugs that block the PGP activity (24).

The efflux transporters are also integral to developmental programs with other roles in development besides toxicant protection. Examples are seen in Dictyostelium, where PGP transporters efflux morphogens signaling stalk cell differentiation (25) and in Arabidopsis, where MRP transporters direct root development by efflux-mediated establishment of auxin gradients (26). In sea urchin embryos, a major MRP activity appears to be involved in regulation of cell-cycle progression, possibly through efflux of an endogenous compound(s) during early cleavages (27). In Caenorhabditis elegans, an MRP transporter is involved in the developmental responses leading to the formation of the dauer larvae, an alternative developmental path chosen when the environment is stressful (28).

A major concern about efficacy of these transporters as a defense of embryos in a changing world is that some drugs or industrial chemicals have structures that are not recognized by the efflux transporters. Thalidomide, the potent teratogen noted earlier, is one example of this problem; recent research shows that this drug is not recognized by human PGP (29) and therefore cannot be kept out of the embryo by this defense mechanism.

Antioxidant Strategies.

Oxidative stress is another formidable problem during development. Metals, ionizing radiation, and various toxicants catalyze formation of reactive oxidative species, and potent oxidants are also produced by embryo metabolism. For example, fertilization of sea urchin eggs results in production of lipid peroxides (30), nitric oxide (31), and hydrogen peroxide (32). The hydrogen peroxide (32) is involved in elevation and hardening of the fertilization envelope, which is itself a protective structure. One defensive mechanism against endogenous production of peroxide includes degradation within the cell by the action of the dual oxidase that produces it.

Another antioxidant mechanism used in sea urchin eggs comes from ovothiol, an embryo-specific mercaptohistidine compound that acts as a more potent antioxidant than glutathione (33). In the sea urchin embryo, glutathione and ovothiol are present at a total concentration of 10 mM, and relative levels of oxidized and reduced glutathione remain constant despite the endogenous production of oxidants at fertilization. In these embryos, these constant glutathione levels are maintained by activation of the pentose phosphate pathway at fertilization, which maintains high NADPH used for recycling of oxidized thiols back to their reduced state (34). An additional bolster for constant redox level is the activation of an NAD kinase at fertilization, which triples the amount of NADPH in the cell.

In mice, the unfertilized egg contains 7 mM glutathione. However, unlike those in the sea urchin embryo, glutathione levels are not stable, and the reduced form of this compound (GSH) decreases 10-fold through preimplantation development to the blastocyst stage (35). Xenobiotics that cause oxidative stress speed this depletion of reduced glutathione and overwhelm this defense. For example, the damage from thalidomide most likely ensues from its oxidation by prostaglandin synthase (36) and concomitant production of reactive oxygen species. Rabbit embryos, which are relatively susceptible to thalidomide, deplete glutathione to 50% of control levels when exposed to 15 μM concentrations of this drug (37). Recent research suggests that the limb malformations characteristic of thalidomide are caused by interference with NF-κB signaling in developing limb buds in response to the altered oxidative status (38).

Protein Protection.

Heat and other stressors can denature intracellular proteins. The major protection is provided by heat-shock proteins (HSPs), molecular chaperones that stabilize the hydrophobic regions of proteins exposed during protein synthesis, translocation across membranes, or after exposure to protein-denaturing stressors or toxicants (39, 40).

The general strategy in adult cells is for HSPs to be constitutively expressed at low levels and synthesis to be up-regulated in response to protein-denaturing stressors. In contrast, HSPs are not induced in response to stress during early cleavage stages of the orphan embryos (41). This strategy may be in part because overexpression of these proteins in embryos inhibits development (42), possibly by interfering with mitosis or signaling (43, 44). Instead the strategy is to have adequate, but not excessive, HSP present in the egg at fertilization. For example, Drosophila, which encounter high heat during normal development in rotting fruit, have sufficient HSP70 levels to handle this stress.

In later development most embryos can up-regulate the levels of existing HSPs or synthesize new types of HSPs. These newly synthesized proteins can serve developmental roles, as in HSP-mediated regulation of lens differentiation in vertebrates (45), or the HSPs can be made for protection and repair when embryos encounter environmental stress. Finally, as we describe later, HSPs such as HSP26 are also synthesized by embryos to stabilize cells during adaptation to extreme stress.

Buffering and Robustness

There is also an intrinsic robustness of developmental programs that insures that development proceeds normally in the face of stresses that could disrupt cellular and molecular mechanisms involved in embryogenesis. This robustness, or buffering, is evidenced by developmental fidelity despite environmental perturbation. For example, maintenance of the anterior–posterior axis in Drosophila is determined by a gradient of the morphogen bicoid, and the position of the throracic segments is defined by this gradient (46, 47). If fly embryos are raised under different temperatures the bicoid boundary broadens, which should seemingly alter segment position; yet the segment positions are still established accurately (48). This robustness is, in part, mediated by some as yet unknown action of the RNA-binding protein staufen (48).

Stability is an inherent feature of many developmental gene networks (49). Models of these networks, such as the segmentation and neurogenic networks of flies, indicate that these processes are stable despite simulated variation in the levels of multiple genes within the networks (50, 51). This stability appears to be provided by the architecture of the network itself and the presence of multiple levels of control of gene expression. For example, the fly neurogenic network appears stabilized by the multiple levels of negative feedback provided by transcription factors in the “enhancer of split” complex.

Although genetic networks are robust, recent research suggests that some diseases can be traced to specific susceptibilities in these networks. For example, analysis of gene activity in trisomy in a mouse model of Down's syndrome indicates that the 1.5-times increased concentration of two genes, DSCR and DYRK, disrupts the NFAT signaling network. Transgenic mice with 2- to 3-fold higher levels of these genes show many of the cardiovascular defects characteristic of human trisomy as consequence of this disruption of the NFAT network (52).

Alternative Developmental Pathways

Another approach to dealing with environmental stresses, used by embryos of many species, is to have intrinsic programs that can alter developmental path and resultant phenotype in response to the environment encountered. There two types of developmental “decisions” used by embryos: emergency responses and adaptive tuning. Emergency responses, including facultative diapause and delay, cope with poor conditions encountered during development, such as desiccation or low food availability, that are so severe that development will be compromised. Adaptive tuning is a response to expected or anticipated changes in the environment and ensue from developmental plasticity (reviewed in refs. 53 and 54).

Emergency Responses.

Embryos of many species temporarily slow or suspend development under severe environmental conditions. In embryos of some mammals, for example, the duration of gestation is controlled by availability of food. If availability is low, development of the embryo slows or arrests, usually at the blastocyst stage (55). A similar example is seen in Lymnea. Embryos of this snail will slow the tempo of development when the pond is crowded and embryos are food-limited. This response is triggered by a neurotransmitter released by juveniles living in the same pond; the signal is sensed by larval chemosensory neurons, and the response is slowing of developmental rate (56).

Embryos of the brine shrimp Artemia (57), killifish (58), and nematodes (59) have a more dramatic response. These embryos, which normally develop directly to the adult stage under nonstress conditions, will, under stressful conditions, dramatically alter the developmental program. For instance, Artemia embryos can completely arrest in a metabolically suppressed diapause gastrula stage. Remarkably, these “encysted” embryos can survive years of continuous anoxia (60), desiccation, and UV radiation (57). Major adaptations include massive accumulation of the sugar trehalose, which helps offset desiccation, and the synthesis and accumulation in the nucleus of an HSP (p26), which is thought to protect the nuclear proteins during these extreme conditions (61).

A different developmental approach is seen in C. elegans. These embryos normally progress through four larval stages before reaching the reproductive adult form. If they encounter stressful conditions, particularly food limitation, the third-stage larva will be redirected to instead form a resistant dauer stage that is nonfeeding and can survive for several months. Unlike brine shrimp or killifish, this larva has measurable metabolic activity. Like the snail Lymnea, the signal of a stressful environment is a pheromone released by other juvenile or adult worms living in the same area. Also like Lymnea, the dauer pheromone (62) is sensed by chemosensory neurons (63) but is here transduced to redirect development to the dauer form.

Adaptive Tuning.

This form of developmental decision results in a different adult phenotype in response to the environment encountered as an embryo (reviewed in refs. 2, 53, and 54). This adaptive variation, also termed predictive adaptive response, produces offspring that are best suited for the environment they are likely to encounter later, as juveniles or adults. In a classic example, butterfly embryos and larvae develop a conspicuous eyespot in the wet season when cover will be abundant, whereas the dry-season larvae do not develop the eye spot when cover will be decreasing (64). A recently described example of adaptive plasticity is in Daphnia, where, if healthy mothers encounter infectious bacteria during reproduction, they produce offspring that are more resistant to infection (65). As we will discuss below, this tuning can be maladaptive when there is a mismatch between the actual and anticipated environments.

When Development Goes Awry

We have already alluded to how developmental failure can occur despite the potent defenses in embryos. Teratology seeks to identify the environmental factors that derail development and elucidate the mechanisms by which they act. Recent work reveals that these factors work in several predictable ways, including defeating embryo defenses, interfering with hormones that control developmental decisions, and inducing maladaptive developmental responses.

Bypassing Defenses.

As suggested earlier, an unintended consequence of industrial and pharmaceutical chemistry is the production of teratogenic chemicals that are not recognized or neutralized by the embryo defenses, best illustrated by recent findings on the teratogen thalidomide, a chemical that was instrumental in shaping our perception of the vulnerability of embryos. This work shows that thalidomide is not recognized by the PGP efflux transporter, which normally provides a first line of defense against toxicants crossing the placenta and entering the fetus (29). Because thalidomide is not a substrate for this transporter, the drug has unlimited access to the developing embryo, and its subsequent metabolism within the embryo produces free radicals that overwhelm key antioxidant defenses. One view is that the subsequent changes in intracellular physiology cause abnormal limb development by misdirecting a redox-sensitive developmental signaling pathway (38).

Hormonal Interference.

Another common mechanism of teratology is interfering with processes in development, such as the determination of secondary sexual characteristics, that are products of balanced signaling “seesaws” controlled by hormones. These processes present specific windows of vulnerability to the environment, windows which are uniquely present during development but absent in the adult. Of concern are a group of chemicals, commonly referred to as endocrine-disrupting compounds, that can derail sexual differentiation during development. A recent example of this is seen with exposure to the triazine herbicide atrazine during amphibian development. The consequence of exposure to low doses is the formation of oocytes within the testis of male frogs (66). Atrazine is thought to induce synthesis of aromatase, an enzyme that converts testosterone to estradiol with resultant effects on sex determination during the vulnerable period of gonadal development. In contrast, this window of susceptibility is absent in the sexually differentiated adult frog. Here atrazine alters steroidogenesis through an aromatase-dependent mechanism, as seen in the embryo, but does not change gonadal phenotype (67).

A different type of reproductive effect is seen in rats exposed to methoxychlor, another endocrine-disrupting compound. Exposure during the period of testis determination results in male offspring with reduced spermatogenesis, reduced sperm counts, and lower sperm motility (68). This phenotype persists in subsequent generations, suggesting some sort of inherited epigenetic phenomenon such as DNA methylation. The period of sensitivity to methoxychlor corresponds to the time when changes in DNA methylation are occurring in the primordial sperm stem cells, and, indeed, there is an altered level of DNA methylation in embryos exposed to this chemical in utero. Although a caveat of this work is that high concentrations of methoxychlor were used, the findings raise appropriate and unresolved questions about low-level effects in humans and wildlife.

Maladaptive Developmental Responses.

Another mode of teratology is caused by inappropriate adaptive responses to stress during embryogenesis, and the effects of these maladaptive responses are often seen later in life. A dramatic example comes from epidemiological studies showing that adverse conditions during pregnancy result in increased incidence of chronic disease in adulthood. One likely mechanism for this correlation is that women who encounter adverse nutritional conditions during pregnancy bear children with a “metabolically thrifty” phenotype. This altered phenotype is presumably adapted to more efficient utilization of food. However if these thrifty individuals have high caloric intake and low energy expenditure in adulthood, the mismatch between their physiology (determined while in utero) and environment (encountered as an adult) increases their likelihood of developing such chronic diseases as diabetes and hypertension (69).

Studies on rats and mice confirm the importance of maternal diet on physiology and even behavior of offspring (70, 71). One mechanism that might account for the relationship between maternal diet and adult disease is that maternal nutrition status might induce an epigenetic change, as by DNA methylation or some aspect of chromatin structure. Such alterations in DNA or chromatin change gene expression, inhibiting or promoting activity, and the changes can be long-term and, in some cases, inherited. Further support for this view comes from research on dietary augmentation with vitamins that act as cofactors for methyl transfer, such as folic acid and vitamin B12 (72). This work shows that supplements can increase DNA methylation, which can repress gene expression (73). An alternative view is that these nutritional effects might not be through DNA methylation, but some other epigenetic alteration directly on chromatin structure (74).

Can Seemingly Abnormal Development Have an Adaptive Value?

The induction of phenotypic variation by stress during development may also be part of a bet-hedging strategy in which variants are produced in the anticipation that some of these variants will be adaptive. These variants might only be of single generation advantage in the case of epigenetic changes restricted to somatic cells. If, however, the epigenetic change occurs in the germ line, then that alteration would be transmitted to subsequent generations.

An example of a novel inheritable (75) and possibly epigenetic (76) mechanism that is mediated by stress is seen with stressors that disrupt the chaperone HSP90 during development. This protein can act as a buffer by stabilization of “client” proteins in developmental signaling pathways. HSP90 is unique because its buffering action also prevents expression of genetic variants that become apparent only when HSP90 is overwhelmed. The resulting expression of these previously quiescent mutations results in radically altered phenotypes (77). This generation of new phenotypes is a “lottery approach,” meaning that most of these phenotypes are not adaptive. However, the gamble may provide a chance of escape from severe environmental bottlenecks.

Epigenetic phenomena, whether mediated by alteration of gene expression, as through methylation, or chromatin changes or HSP90 buffering, also highlight how stability and robustness in development are balanced against plasticity and environmentally driven expression of new, possibly adaptive, phenotypes (78). Excessive buffering of development may represent a cost to the organism if it prevents expression of novel phenotypes. The need to maintain this plasticity might explain why embryos are not buffered, or canalized, to the environmental limits for survival (Fig. 2).

Fig. 2.

Fig. 2.

Development is buffered within a range of environmental conditions, around the environmental optimum. As environment deviates from the optimum, the embryo adapts and some adaptive responses can maintain optimal fitness of the embryo. Under more severe conditions teratology results either as a consequence of maladaptive response or overwhelming of the adaptive and protective systems. The shape of this relationship between the embryo and the environment varies according to the species and environmental factor in question. The magnitude of buffering around these optima may be broader for some embryos than others. One view is that most model organisms used in developmental biology have broadly buffered developmental limits (87).

Predictive Value of Understanding Embryo Defenses

Wilson's paradigm predated much of the research we have reviewed here on robustness of the embryo. The general theme seen in these protective strategies is to prepare the embryo for an expected environment(s). This strategy can be ineffective if those environments are rapidly altered and if new, unanticipated stressors are added to the challenges facing the embryo. Understanding these mechanisms of embryo robustness is necessary to predict and prevent instances where these defenses will be overwhelmed. Most of these situations arise from rapid changes, which defeat their adaptive strategies and defenses.

As illustrated in this review, the defenses present in embryos are not always capable of recognizing and neutralizing man-made chemicals such as drugs and pesticides, but understanding the limits and mechanisms of these defenses can provide predictive tools to identify risky compounds. For instance, because efflux transporters are the primary defense that protects the embryo against chemicals, it seems prudent to screen compounds to which embryos could be exposed to determine whether the chemicals are cell-permeable and are substrates of the efflux transporters. Chemicals that are permeable, but not substrates, would be predicted to have unrestricted access to embryos (24).

Another concern relates to the ability of some compounds to inhibit the activity of these transporters and render the cell more permeable to compounds that are otherwise kept out: a phenomenon sometimes referred to as “chemosensitization.” Synthetic fragrances, which are widely used in household products, are one class of compounds recently shown to inhibit these transporters in cells from marine mussels (79). There is an immediate need to find out whether these effects are also seen in mammalian embryos through maternal exposure to these ubiquitous compounds and whether other commonly used substances have similar properties.

Study of these defenses might also provide clues about whether the magnitude and rate of environmental changes will exceed the buffering limits of embryo physiology and the rate of evolution of new defenses. For instance, increased CO2 in the atmosphere is leading to oceanic acidification (80). Recent work indicates that continued acidification would disrupt development of pH-sensitive calcium carbonate skeletons in larvae such as those in sea urchins (81). Similarly, increasing global temperature may become a problem. Studies have shown that adult organisms from the intertidal and coral reef habitats are already living at or near their physiological limits for temperature (82), but there is little information of the thermal buffering limits for most embryos and larvae.

It is also critical to understand the mechanisms by which endocrine-disrupting chemicals misdirect development during critical environmentally sensitive windows. A pertinent example is bisphenol A, a chemical widely used in plastics and having estrogen-like activity, affecting development and perhaps even causing adult onset diseases as a consequence of fetal exposure (83). The continued use of this chemical has been controversial, and a major factor hindering regulation of this chemical is the lack of a clear mechanism to explain the observed effects. If estrogen-like, bisphenol A would presumably act through conventional nuclear estrogen receptors, yet the binding to these receptors is weak.

The solution to this conundrum may come from changing views on signal transduction and from new findings on mechanisms of bisphenol action. We are starting to appreciate that signaling networks respond in combinatorial fashion to agonists such that different agonists that result in seemingly identical intracellular signals, such as changes in calcium or activation of kinase pathways, can result in different outputs in phenotypic response (84). A new finding on the mechanism of action of bisphenol A is that it binds to a group of membrane receptors that also interact with estradiol and other signaling molecules at the very low environmental levels (e.g., 10−9 M) seen in humans. This receptor acts through “traditional” signaling cascades, increasing calcium levels and downstream MAP kinase pathways (85, 86). This description of mode of action at these low concentrations, similar to the levels seen in humans, showcases how fundamental research can lead to results with a potentially large impact on use of chemicals that impact human health.

The formidable challenges to understand and prevent pathologies such as cancer or infectious disease are widely appreciated. Describing and preventing emerging environmental problems for embryos and deciphering the developmental origins of adult disease are problems of equal importance, magnitude, and complexity. Solutions to these problems require a focus on the biology of the embryo within its environmental context and an understanding of the mechanisms that cause development to go awry in a rapidly changing world. The implications are far-reaching for human and global fitness.

Acknowledgments

We thank our many colleagues who read and commented on various versions of this manuscript and especially James Clegg, George von Dassow, Kathy Foltz, Mark Hahn, Jason Hodin, David McClay, John Pearse, George Somero, David Stevenson, and Cathy Thaler. We thank Julia Cardosa and Chris Patton for preparing the illustrations. This work was supported by National Institutes of Health Grant F32-HD47136, National Science Foundation Grant 0446384, the Stanford Environmental Initiative, and the Stanford Neonatology Program.

Abbreviations

HSP

heat-shock protein

MRP

multidrug-resistance protein

UV

ultraviolet radiation.

Footnotes

The authors declare no conflict of interest.

References

  • 1.Wilson JG. Environment and Birth Defects. New York: Academic; 1973. [Google Scholar]
  • 2.Gilbert SF. Dev Biol. 2001;233:1–12. doi: 10.1006/dbio.2001.0210. [DOI] [PubMed] [Google Scholar]
  • 3.Strathmann RR, Staver JM, Hoffman JR. Evol Int J Org Evol. 2002;56:708–720. doi: 10.1111/j.0014-3820.2002.tb01382.x. [DOI] [PubMed] [Google Scholar]
  • 4.Shick JM, Dunlap WC. Annu Rev Physiol. 2002;64:223–262. doi: 10.1146/annurev.physiol.64.081501.155802. [DOI] [PubMed] [Google Scholar]
  • 5.Adams N, Shick M. Mar Biol. 2001;138:267–280. [Google Scholar]
  • 6.Karentz D, Bosch I, Mitchell DM. Mar Biol. 2004;145:277–292. [Google Scholar]
  • 7.Kiesecker JM, Blaustein AR, Belden LK. Nature. 2001;410:681–684. doi: 10.1038/35070552. [DOI] [PubMed] [Google Scholar]
  • 8.Prager EM, Wilson AC, Arnheim N. J Biol Chem. 1974;249:7295–7297. [PubMed] [Google Scholar]
  • 9.Gorman MJ, Kankanala P, Kanost MR. Insect Mol Biol. 2004;13:19–24. doi: 10.1111/j.1365-2583.2004.00454.x. [DOI] [PubMed] [Google Scholar]
  • 10.Kudo S, Teshima C. J Exp Zool. 1991;259:392–398. [Google Scholar]
  • 11.Gil-Turnes MS, Hay ME, Fenical W. Science. 1989;246:116–118. doi: 10.1126/science.2781297. [DOI] [PubMed] [Google Scholar]
  • 12.Kaufman MR, Ikeda Y, Patton C, van Dykhuizen G, Epel D. Biol Bull. 1998;194:36–43. doi: 10.2307/1542511. [DOI] [PubMed] [Google Scholar]
  • 13.Miyake K, McNeil P. Biol Bull. 1998;195:214–215. doi: 10.2307/1542846. [DOI] [PubMed] [Google Scholar]
  • 14.Thomas FIM, Edwards KA, Bolton TF, Sewell MA, Zande JM. Biol Bull. 1999;197:7–10. doi: 10.2307/1542991. [DOI] [PubMed] [Google Scholar]
  • 15.Terasaki M, Miyake K, McNeil PL. J Cell Biol. 1997;139:63–74. doi: 10.1083/jcb.139.1.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.McNeil PL, Terasaki M. Nat Cell Biol. 2001;3:E124–E129. doi: 10.1038/35074652. [DOI] [PubMed] [Google Scholar]
  • 17.Rose RL, Levi PE. In: A Textbook of Modern Toxicology. 3rd Ed. Hodgson E, editor. New York: Wiley; 2004. pp. 149–161. [Google Scholar]
  • 18.Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman MM. Annu Rev Pharmocol Toxicol. 1999;39:361–398. doi: 10.1146/annurev.pharmtox.39.1.361. [DOI] [PubMed] [Google Scholar]
  • 19.Cole SPC, Deeley RG. BioEssays. 1998;20:931–940. doi: 10.1002/(SICI)1521-1878(199811)20:11<931::AID-BIES8>3.0.CO;2-J. [DOI] [PubMed] [Google Scholar]
  • 20.Sarkadi B, Ozvegy-Laczka C, Nemet K, Varadi A. FEBS Lett. 2004;567:116–120. doi: 10.1016/j.febslet.2004.03.123. [DOI] [PubMed] [Google Scholar]
  • 21.Toomey BH, Epel D. Biol Bull. 1993;185:355–364. doi: 10.2307/1542476. [DOI] [PubMed] [Google Scholar]
  • 22.Leazer TM, Klaassen CD. Drug Metab Dispos. 2003;31:153–167. doi: 10.1124/dmd.31.2.153. [DOI] [PubMed] [Google Scholar]
  • 23.Elbling L, Berger W, Rehberger A, Waldhor T, Micksche M. FASEB J. 1993;7:1499–1506. doi: 10.1096/fasebj.7.15.7903262. [DOI] [PubMed] [Google Scholar]
  • 24.Smit JW, Huisman MT, van Tellingen O, Wiltshire HR, Schinkel AH. J Clin Invest. 1999;104:1441–1447. doi: 10.1172/JCI7963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Good JR, Kuspa A. Dev Biol. 2000;220:53–61. doi: 10.1006/dbio.2000.9611. [DOI] [PubMed] [Google Scholar]
  • 26.Gaedeke N, Klein M, Kolukisaoglu U, Forestier C, Muller A, Ansorge M, Becker D, Mamnun Y, Kuchler K, Schulz B, et al. EMBO J. 2001;20:1875–1887. doi: 10.1093/emboj/20.8.1875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hamdoun AM, Cherr GN, Roepke TA, Epel D. Dev Biol. 2004;276:452–462. doi: 10.1016/j.ydbio.2004.09.013. [DOI] [PubMed] [Google Scholar]
  • 28.Yabe T, Suzuki N, Furukawa T, Ishihara T, Katsura I. Development (Cambridge, UK) 2005;132:3197–3207. doi: 10.1242/dev.01909. [DOI] [PubMed] [Google Scholar]
  • 29.Zimmermann C, Gutmann H, Drewe J. Cancer Chemother Pharmacol. 2006;57:599–606. doi: 10.1007/s00280-005-0087-3. [DOI] [PubMed] [Google Scholar]
  • 30.Perry G, Epel D. Dev Biol. 1985;107:58–65. doi: 10.1016/0012-1606(85)90375-6. [DOI] [PubMed] [Google Scholar]
  • 31.Thaler CD, Epel D. Curr Pharm Des. 2003;9:399–409. doi: 10.2174/1381612033391748. [DOI] [PubMed] [Google Scholar]
  • 32.Wong JL, Creton R, Wessel GM. Dev Cell. 2004;7:801–814. doi: 10.1016/j.devcel.2004.10.014. [DOI] [PubMed] [Google Scholar]
  • 33.Shapiro BM. Science. 1991;252:533–536. doi: 10.1126/science.1850548. [DOI] [PubMed] [Google Scholar]
  • 34.Schomer B, Epel D. Dev Biol. 1998;203:1–11. doi: 10.1006/dbio.1998.9044. [DOI] [PubMed] [Google Scholar]
  • 35.Gardiner CS, Reed DJ. Biol Reprod. 1994;51:1307–1314. doi: 10.1095/biolreprod51.6.1307. [DOI] [PubMed] [Google Scholar]
  • 36.Parman T, Wiley MJ, Wells PG. Nat Med. 1999;5:582–585. doi: 10.1038/8466. [DOI] [PubMed] [Google Scholar]
  • 37.Hansen JM, Carney EW, Harris C. Reprod Toxicol. 1999;13:547–554. doi: 10.1016/s0890-6238(99)00053-2. [DOI] [PubMed] [Google Scholar]
  • 38.Hansen JM, Harris C. Antioxid Redox Signal. 2004;6:1–14. doi: 10.1089/152308604771978291. [DOI] [PubMed] [Google Scholar]
  • 39.Hartl FU, Hayer-Hartl M. Science. 2002;295:1852–1858. doi: 10.1126/science.1068408. [DOI] [PubMed] [Google Scholar]
  • 40.Morimoto R, Tissieres A, Georgopoulos C. The Biology of Heat Shock Proteins and Molecular Chaperones. Plainview, NY: Cold Spring Harbor Lab Press; 1994. [Google Scholar]
  • 41.Heikkila JJ. Semin Cell Dev Biol. 2003;14:259–266. doi: 10.1016/j.semcdb.2003.09.022. [DOI] [PubMed] [Google Scholar]
  • 42.Krebs RA, Feder ME. J Insect Physiol. 1998;44:1091–1101. doi: 10.1016/s0022-1910(98)00059-6. [DOI] [PubMed] [Google Scholar]
  • 43.Nollen EA, Morimoto RI. J Cell Sci. 2002;115:2809–2816. doi: 10.1242/jcs.115.14.2809. [DOI] [PubMed] [Google Scholar]
  • 44.Isaenko OA, Karr TL, Feder ME. Cell Stress Chaperones. 2002;7:297–308. doi: 10.1379/1466-1268(2002)007<0297:hatpmd>2.0.co;2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Evans TG, Yamamoto Y, Jeffery WR, Krone PH. Cell Stress Chaperones. 2005;10:66–78. doi: 10.1379/CSC-79R.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Driever W, Nusslein-Volhard C. Cell. 1988;54:95–104. doi: 10.1016/0092-8674(88)90183-3. [DOI] [PubMed] [Google Scholar]
  • 47.Driever W, Nusslein-Volhard C. Cell. 1988;54:83–93. doi: 10.1016/0092-8674(88)90182-1. [DOI] [PubMed] [Google Scholar]
  • 48.Houchmandzadeh B, Wieschaus E, Leibler S. Nature. 2002;415:798–802. doi: 10.1038/415798a. [DOI] [PubMed] [Google Scholar]
  • 49.Davidson EH, Rast JP, Oliveri P, Ransick A, Calestani C, Yuh CH, Minokawa T, Amore G, Hinman V, Arenas-Mena C, et al. Science. 2002;295:1669–1678. doi: 10.1126/science.1069883. [DOI] [PubMed] [Google Scholar]
  • 50.Meir E, von Dassow G, Munro E, Odell GM. Curr Biol. 2002;12:778–786. doi: 10.1016/s0960-9822(02)00839-4. [DOI] [PubMed] [Google Scholar]
  • 51.von Dassow G, Meir E, Munro EM, Odell GM. Nature. 2000;406:188–192. doi: 10.1038/35018085. [DOI] [PubMed] [Google Scholar]
  • 52.Arron JR, Winslow MM, Polleri A, Chang CP, Wu H, Gao X, Neilson JR, Chen L, Heit JJ, Kim SK, et al. Nature. 2006;441:595–600. doi: 10.1038/nature04678. [DOI] [PubMed] [Google Scholar]
  • 53.van der Weele C. Images of Development: Environmental Causes in Ontegeny. Albany, NY: State Univ of New York; 1999. [Google Scholar]
  • 54.West-Eberhard MJ. Developmental Plasticity and Evolution. New York: Oxford Univ Press; 2003. [Google Scholar]
  • 55.Renfree MB, Shaw G. Annu Rev Physiol. 2000;62:353–375. doi: 10.1146/annurev.physiol.62.1.353. [DOI] [PubMed] [Google Scholar]
  • 56.Voronezhskaya EE, Khabarova MY, Nezlin LP. Development (Cambridge, UK) 2004;131:3671–3680. doi: 10.1242/dev.01237. [DOI] [PubMed] [Google Scholar]
  • 57.Tanguay JA, Reyes RC, Clegg JS. J Biosci. 2004;29:489–501. doi: 10.1007/BF02712121. [DOI] [PubMed] [Google Scholar]
  • 58.Wourms JP. J Exp Zool. 1972;182:389–414. doi: 10.1002/jez.1401820310. [DOI] [PubMed] [Google Scholar]
  • 59.Cassada RC, Russell RL. Dev Biol. 1975;46:326–342. doi: 10.1016/0012-1606(75)90109-8. [DOI] [PubMed] [Google Scholar]
  • 60.Clegg J. J Exp Biol. 1997;200:467–475. doi: 10.1242/jeb.200.3.467. [DOI] [PubMed] [Google Scholar]
  • 61.Willsie JK, Clegg JS. J Cell Biochem. 2002;84:601–614. [PubMed] [Google Scholar]
  • 62.Jeong PY, Jung M, Yim YH, Kim H, Park M, Hong E, Lee W, Kim YH, Kim K, Paik YK. Nature. 2005;433:541–545. doi: 10.1038/nature03201. [DOI] [PubMed] [Google Scholar]
  • 63.Bargmann CI, Horvitz HR. Science. 1991;251:1243–1246. doi: 10.1126/science.2006412. [DOI] [PubMed] [Google Scholar]
  • 64.Brakefield PM, Reitsma N. Ecological Entymology. 1991;16:291–303. [Google Scholar]
  • 65.Mitchell SE, Read AF. Proc Biol Sci. 2005;272:2601–2607. doi: 10.1098/rspb.2005.3253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Hayes T, Haston K, Tsui M, Hoang A, Haeffele C, Vonk A. Environ Health Perspect. 2003;111:568–575. doi: 10.1289/ehp.5932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Hayes TB, Collins A, Lee M, Mendoza M, Noriega N, Stuart AA, Vonk A. Proc Natl Acad Sci USA. 2002;99:5476–5480. doi: 10.1073/pnas.082121499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Anway MD, Cupp AS, Uzumcu M, Skinner MK. Science. 2005;308:1466–1469. doi: 10.1126/science.1108190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Gluckman PD, Hanson MA. Science. 2004;305:1733–1736. doi: 10.1126/science.1095292. [DOI] [PubMed] [Google Scholar]
  • 70.Ecker DJ, Stein P, Xu Z, Williams CJ, Kopf GS, Bilker WB, Abel T, Schultz RM. Proc Natl Acad Sci USA. 2004;101:1595–1600. doi: 10.1073/pnas.0306846101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Keen CL, Hanna LA, Lanoue L, Uriu-Adams JY, Rucker RB, Clegg MS. J Nutr. 2003;133:1477S–1480S. doi: 10.1093/jn/133.5.1477S. [DOI] [PubMed] [Google Scholar]
  • 72.Friso S, Choi SW, Girelli D, Mason JB, Dolnikowski GG, Bagley PJ, Olivieri O, Jacques PF, Rosenberg IH, Corrocher R, Selhub J. Proc Natl Acad Sci USA. 2002;99:5606–5611. doi: 10.1073/pnas.062066299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Robertson KD, Wolffe AP. Nat Rev Genet. 2000;1:11–19. doi: 10.1038/35049533. [DOI] [PubMed] [Google Scholar]
  • 74.Blewitt ME, Vickaryous NK, Paldi A, Koseki H, Whitelaw E. PLoS Genet. 2006;2:e49. doi: 10.1371/journal.pgen.0020049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Rutherford SL, Lindquist S. Nature. 1998;396:336–342. doi: 10.1038/24550. [DOI] [PubMed] [Google Scholar]
  • 76.Sollars V, Lu X, Xiao L, Wang X, Garfinkel MD, Ruden DM. Nat Genet. 2003;33:70–74. doi: 10.1038/ng1067. [DOI] [PubMed] [Google Scholar]
  • 77.Sangster TA, Lindquist S, Queitsch C. BioEssays. 2004;26:348–362. doi: 10.1002/bies.20020. [DOI] [PubMed] [Google Scholar]
  • 78.Kirschner M, Gerhart J. Proc Natl Acad Sci USA. 1998;95:8420–8427. doi: 10.1073/pnas.95.15.8420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.Luckenbach T, Epel D. Environ Health Perspect. 2005;113:17–24. doi: 10.1289/ehp.7301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80.Caldeira K, Wickett ME. Nature. 2003;425:365. doi: 10.1038/425365a. [DOI] [PubMed] [Google Scholar]
  • 81.Kurihara H, Shirayam Y. Mar Ecol Prog Ser. 2004;274:161–169. [Google Scholar]
  • 82.Stillman JH. Science. 2003;301:65. doi: 10.1126/science.1083073. [DOI] [PubMed] [Google Scholar]
  • 83.Welshons WV, Nagel SC, vom Saal FS. Endocrinology. 2006;147:S56–S69. doi: 10.1210/en.2005-1159. [DOI] [PubMed] [Google Scholar]
  • 84.Cancela JM, Van Coppenolle F, Galione A, Tepikin AV, Petersen OH. EMBO J. 2002;21:909–919. doi: 10.1093/emboj/21.5.909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 85.Nadal A, Ropero AB, Fuentes E, Soria B, Ripoll C. Steroids. 2004;69:531–536. doi: 10.1016/j.steroids.2004.05.010. [DOI] [PubMed] [Google Scholar]
  • 86.Nadal A, Ropero AB, Fuentes E, Soria B. Trends Pharmacol Sci. 2001;22:597–599. doi: 10.1016/s0165-6147(00)01846-0. [DOI] [PubMed] [Google Scholar]
  • 87.Bolker JA. BioEssays. 1995;17:451–455. doi: 10.1002/bies.950170513. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES