The human placenta – An alternative for studying foetal exposure

Abstract

Pregnant women are daily exposed to a wide selection of foreign substances. Sources are as different as lifestyle factors (smoking, daily care products, alcohol consumption, etc.), maternal medication or occupational/environmental exposures. The placenta provides the link between mother and foetus, and though its main task is to act as a barrier and transport nutrients and oxygen to the foetus, many foreign compounds are transported across the placenta to some degree and may therefore influence the unborn child. Foetal exposures to environmental and medicinal products may have impact on the growth of the foetus (e.g. cigarette smoke) and development of the foetal organs (e.g. methylmercury and thalidomide). The scope of this review is to give insight to the placental anatomy, development and function. Furthermore, the compounds physical properties and the transfer mechanism across the placental barrier are evaluated. In order to determine the actual foetal risk from exposure to a chemical many studies regarding the topic are necessary, including means of transportation, toxicological targets and effects. For this purpose several in vivo and in vitro models including the placental perfusion system are models of choice.

Introduction

During the last 50 years evidence has accumulated that many pharmacologic agents and also environmental pollutants from surroundings are transferred from the mother to the embryo or foetus. Regardless of whether this transfer is on purpose as in a medical treatment or unwanted as in women unaware of early pregnancy taking over-the-counter drugs as self-medication, more attention and information for the public is needed (Plonait and Nau, 2004). The first evidence of reproductive toxicity caused by a foetal exposure from maternal intake was the thalidomide disaster in 1957–1961. Pregnant women from approximately 46 countries worldwide were prescribed thalidomide as a safe anti-emetic sedative and anti-anxiety drug. Given between the 34th and 50th day of pregnancy thalidomide can exert teratogenic effects seen as skeletal malformation of especially the limbs in 10,000 surviving babies. When further investigated the mechanism of action of thalidomide was specific in humans but confirmed in animal testing and therefore the demands on testing of drugs to be used in pregnancy were increased to include two or more animal species (Botting, 2002, Brent, 2004). Later, in the early 1970s it was scientifically demonstrated that prenatal alcohol exposure can cause mental retardation, facial malformations, prenatal and/or postnatal growth retardation (West and Blake, 2005, Riley and McGee, 2005). In 1971 the reproductive damaging effect of the synthetic nonsteroidal estrogen diethylstilbestrol (DES) became evident. The drug was prescribed to prevent miscarriage and other pregnancy complications but as an unknown teratogenic effect, it also caused carcinomas in vagina and cervix in young women offspring and malformation of reproductive organs in both girl and boy offspring (Swan, 2000). This led to a broader definition of reproductive damaging effects, including not only functional and cognitive effects seen at birth, but also effects seen later in life caused by a foetal exposure. It is now common knowledge that maternal smoking (Habek et al., 2002), and exposure to methylmercury, lead, environmental chemicals as polychlorinated biphenyls (Schantz, 1996, Tilson et al., 1998) may cause negative health effects in the human offspring. The scientific communities and public now pay more attention to the potential teratogenic and foetotoxic effects with increased focus on foetal exposure.

In this review, we will give a short introduction to the placenta and its function. Furthermore, the mechanisms allowing drugs and substances to cross the placental barrier will be described, and different methods to investigate the placental transport and toxicity of endogenous compounds. Emphasis is on the placental perfusion system as it is an interesting tool for current studies of drug transfer.

The human placenta is unique in structure (anatomy, pathology, and physiology) and only resembles placenta from certain primate species, e.g. the macaque (Enders and Blankenship, 1999).

The placenta is defined as the fusion of foetal membranes with the uterine mucosa for the purpose of maternofetal exchange of nutrients, gases, and waste substrates. According to this definition, placental development starts at day 6 or 7 post-conception, as soon as the blastocyst starts invasion of the endometrium. The blastocyst consists of an outer cover, the trophoblast, and an inner cell mass, the embryoblast (Kaufmann and Frank, 2004). The trophoblast cell lineage originates from the trophectoderm at the blastocyt stage and the stromal and vascular components of the placenta are derived from the allantois, also of fetal/blastocyst origin (Cross, 2006). It is parts of the trophoblast of the blastocyst that adheres to and invades the uterine mucosa and in turn differentiates to the syncytiotrophoblast. The primary drug transport site in the grown placenta, the villi, is developed during the lacunar period (day 8–13 post-conception). The villous system consisting of trabeculae (separating outgrowths becoming pre-villi) and lacunae (vacuoles) arise in the syncytiotrophoblast. Surrounding this system is the primary chorionic plate (towards the embryoblast) and the trophoblastic shell (towards the endometrium). The pre-villi are formed by the cytotrophoblast that invades the trabeculae from the primary chorionic plate and proliferates inside of these resulting in branches that protrude into the vacuoles. If the villi are attached to the trophoblastic shell (basic plate) they are called anchoring villi. The still expanding lacunae system becomes the intervillous space. The first primitive maternal circulation is established by cells from the trophoblastic shell which manoeuvre themselves into the maternal endometrial vessels (from day 12 post-conception). The maternal blood enters the lacunar system through small holes in the shell (Syme et al., 2004, Kaufmann and Frank, 2004). From the chorionic plate mesenchymal cells invade the primary villi thereby making it secondary villi. Some of these invasive mesenchymal cells are differentiated into hemangioblastic cell cords which further differentiate into the first foetal capillaries (tertiary villi). A population of the same hemangioblastic cell cords differentiates into hematopoietic stem cells which start the blood formation inside the capillaries. At the same time the fetally vascularized allantois reaches the chorionic plate and extends through the plate into the villi where they fuse with the intravillous capillary bed. The result is the intraplacental foetal circulation which is fully established at the end of the fifth week post-conception (Kaufmann and Frank, 2004). However, the complete foetal–placental–maternal circulation is not entirely established until around the tenth week of pregnancy, therefore substances present in the maternal blood until this time must be introduced to the embryo via diffusion through the extracellular fluid (Syme et al., 2004). The foetal circulation ends in the villous trees and these are found in the vascular units (cotyledons) within the placenta. The full-term placenta contain between 10 and 40 cotyledons separated from each other by the placental septa.

The Grosser classification is still widely used as a mean to characterize placenta. Placentae described by the absence of maternal tissue such that maternal blood directly contacts the trophoblast are called haemochorial (humans, monkeys). A schematic representation of the full-term haemochorial placenta is given in Fig. 1. The placental barrier in this type is composed of tissue from foetal origin only. Placentae called endotheliochorial (cat, dog) have brought the trophoblast into contact with the maternal capillary endothelium due to erosion. Epitheliochorial (horse, pig) has six layers of tissue separating the foetal and maternal circulations. Finally, in syndesmochorial (ruminants) placentae the maternal endothelium disappears, resulting in trophoblastic cells in direct contact with maternal connective tissue (Grosser, 1909, Faber et al., 1992, Leiser and Kaufmann, 1994, Simone et al., 1994). During human pregnancy, the foetal and maternal circulations are separated by the placental barrier that consists of five layers in the first trimester: the syncytiotrophoblast layer (lining the villi), the cytotrophoblast, the trophoblastic basal lamina, connective tissue, and the foetal endothelium. This barrier undergoes drastic changes throughout pregnancy: the syncytiotrophoblast layer is largely reduced in thickness, the cytotrophoblast becomes discontinuous, changes in the villus structure are also found; it enlarges probably to ease the exchange-processes between mother and foetus (Fox, 1991, van der Aa et al., 1998, Kaufmann and Frank, 2004). Both the differences in trophoblastic layers and the anatomic differences of the villous trees between species result in a large variation in placental function. Factors as diffusion, electrical potential across the barrier, magnitude of maternal and foetal blood flows, and the possible differences in presence of carriers should be considered. Consequently, placental function between species is different, especially the transfer and metabolism of drugs vary considerably (Simone et al., 1994, Syme et al., 2004).

The placental function varies during pregnancy, thus during early gestation the primary function of the early placenta is to mediate implantation of the embryo into the uterus and, the secondary function is to produce hormones that prevent the end of the ovarian cycle. After implantation, the primary placental function is to regulate nutrient and oxygen uptake from the mother to the foetus. The placenta plays an active role in regulating maternal physiology to the nutritional benefit of the foetus, e.g. the trophoblast produce angiogenic factors and vasodilators, produce hormones that stimulate the maternal blood cell production and blood volume, produce growth hormones and placental lactogens and hormones that suppress and stimulate appetite, etc. (Cross, 2006). These substances and other drugs are transported across the placenta via specific transport mechanisms present in the maternal-facing apical (brush border) membrane and fetal-facing basal membrane of the syncytiotrophoblast (Syme et al., 2004).

It is well established that the placenta metabolizes and transfers a large diversity of pharmacologically active molecules. Therefore, some concern rests in if the placental metabolism converts precursors into potential toxic metabolites. Though, the metabolizing system has metabolic activities and a substrate spectrum seemingly somewhat reduced when compared to the liver, the placenta contains a rich enzymatic machinery able to carry out both Phase I and II reactions (Pasanen and Pelkonen, 1994, Pasanen, 1999, Marin et al., 2004). Several cytochrome P450 proteins including CYP1, CYP2 and CYP3 have been isolated from the placenta. These proteins are largely responsible for the mechanisms in the detoxification of drugs and toxins. The enzymatic machinery has been thoroughly reviewed in Pasanen and Pelkonen, 1994, Marin et al., 2004.

The speed and the extent of compound-transfer depend on the physiochemical and structural characteristics of the drug as well as the physical characteristics of the maternal–placental–embryonic-foetal unit. The physiochemical characteristics of a compound can determine their transfer rate through the placenta, considering the weight, ionization and lipid-solubility of the compound. Molecules with a weight up to 600 Da, non-ionized and lipid soluble will show unimpeded diffusion. The transfer rate is called flow-limited transfer and will depend only on the factors regulating maternal and foetal blood flows. Larger, ionized, hydrophilic compounds will cross the placenta more slowly. Transfer of this type of compounds is referred to as membrane limited transfer and the rate is slower than the blood stream. The constituents of the membranes determined the transfer rate (van der Aa et al., 1998). Hydrophilic molecules encounter a resistance by the haemocordial type of placenta due to the trophoblast and the endothelium (Thornburg and Faber, 1977, van der Aa et al., 1998). The physical factors include the surface area of the exchange membrane; the thickness of the endothelio-syncytial membrane; the maternal blood flow and the hydrostatic pressure in the intervillous chamber; the blood pressure in foetal capillaries and the difference in the osmotic pressure between mother and foetal compartments (Bourget et al., 1995). Closer to term-pregnancy the exchange between mother and child intensifies partly because of the thinning of membranes as described above.

The transfer of compounds can occur by four kinds of mechanisms. First, many compounds will diffuse across the human placenta by the passive diffusion process. This process is a transfer without the use of energy. It is dependent only by the factors introduced in Fick’s law of passive diffusion: rate of diffusion = D × Δc × A/d; the surface area (A); thickness of the membrane barrier (d); the drug concentration gradient across the membrane (between the maternal and fetal blood) (Δc) and the substance specific diffusion constant (D) (Plonait and Nau, 2004). Importantly, the compound characteristics and protein binding capacities also influence the substance’s capability of crossing the placenta. Many pharmacologically active compounds cross the placenta by simple diffusion, although also larger molecules as antibodies may be transported this way (Malek et al., 1998). Second, the transport of the substance is carrier mediated down a concentration gradient without energy-costs called facilitated diffusion. Only a few drugs have been suggested to be transported this way, one of them is ganciclovir against intrauterine infection. Ganciclovir was shown to be taken up by the maternal-facing placental membrane by a carrier-dependent, Na-independent system (Henderson et al., 1993). Generally, compounds structurally related to endogenous compounds intended for this kind of transport, are assumed to use facilitated diffusion, e.g. hormones and nucleosides (van der Aa et al., 1998, Syme et al., 2004). Third, active transport includes the movement of a substance against a chemical or electrical gradient with energy costs. The transport is carrier mediated and there is a high degree of competition between related compounds. An example is the sodium/multivitamin transporter (SMVT) located in the placental brush-border membrane. It has been discussed whether the drugs like carbamazepine compete with endogenous biotin for the SMVT transporter (Ganapathy et al., 2000). Fourth, pinocytosis in which the compound is invaginated into the cell membrane where it is transferred to the opposite site as a vesicle. However, the current conclusion of many placental transport studies is that this process is too slow to be highly relevant for the transfer of drugs from mother to foetus (Syme et al., 2004).

Up to now, approximately 20 different drug transport proteins have been determined (Unadkat et al., 2004, Myllynen et al., 2005). However, it is not clear whether these transporters are coupled to transport of foreign chemical substances – thus they do have the potential. One example is the antiepileptic drug gabapentin that inhibited LAT1-mediated [¹⁴C]phenylalanine uptake in a competitive manner (Uchino et al., 2002, Myllynen et al., 2005).