Yolk sac development, function and role in rodent pregnancy

During the early phases of embryonic development, the yolk sac serves as an initial placenta in many animal species. While in some, this role subsides around the end of active organogenesis, it continues to have important functions in rodents, alongside the chorio‐allantoic placenta. The yolk sac is the initial site of hematopoiesis in many animal species including primates. Cells of epiblastic origin form blood islands that are the forerunners of hematopoietic cells and of the primitive endothelial cells that form the vitelline circulation. The yolk sac is also a major route of embryonic and fetal nutrition apparently as long as it functions. In mammals and especially rodents, macro and micronutrients are absorbed by active pinocytosis into the visceral yolk sac, degraded and the degradation products (i.e., amino acids) are then transferred to the embryo. Interference with the yolk sac function may directly reflect on embryonic growth and development, inducing congenital malformations or in extreme damage, causing embryonic and fetal death. In rodents, many agents were found to damage the yolk sac (i.e., anti–yolk sac antibodies or toxic substances interfering with yolk sac pinocytosis) subsequently affecting the embryo/fetus. Often, the damage to the yolk sac is transient while embryonic damage persists. In humans, decreased yolk sac diameter was associated with diabetic pregnancies and increased diameter was associated with pregnancy loss. In addition, culture of rat yolk sacs in serum obtained from pregnant diabetic women or from women with autoimmune diseases induced severe damage to the visceral yolk sac epithelium and embryonic malformations. It can be concluded that as a result of the crucial role of the yolk sac in the well‐being of the early embryo, any damage to its normal function may severely and irreversibly affect further development of the embryo/fetus.


| INTRODUCTION
One of the favorite topics of Dr. Brent's research interests was the yolk sac, especially in mice and rats. About 50 years ago he felt that this was a neglected area of investigation. It was then already known that the yolk sac had very important biological functions in rodents, is crucial for normal embryonic development and persists throughout the pregnancy following implantation. Moreover, it is the sole functional placenta for more than half of pregnancy in mice and in rats and throughout active organogenesis. Hence, he believed that hindering the function of the yolk sac may play an important role in the etiology and pathogenesis of many congenital malformations.
This is what Dr. Brent and associates wrote in their review published in Teratology in 1990: (Brent et al., 1990) "The visceral yolk sac is involved with nutritional, endocrine, metabolic, immunologic, secretory, excretory and hematopoietic functions". About two decades earlier he also published his studies on the teratogenic effects of anti-yolk sac antibodies demonstrating that interference with yolk sac function may irreversibly obstruct normal embryonic development (Brent, Johnson, & Jensen, 1971). While embryonic damage is final, the yolk sac damage tends to vanish within 2-3 days (Beckman, Ornoy, Jensen, Arnon, & Brent, 1991).
There are plenty of data demonstrating that damage to a functioning yolk sac might negatively affect embryonic development especially in rodents as the visceral yolk sac has important functions almost throughout the entire pregnancy (Brent, Beckman, Jensen, & Koszalka, 1990). However, little is known about the possible effects of embryonic and fetal mal-development on the structure and function of the yolk sac. During the last years, more human data are accumulating demonstrating that embryonic/fetal death or maternal diseases (i.e., diabetes mellitus) might affect the yolk sac, inducing changes in its size and shape (Jaiman, Romero, Pacora, et al., 2021;Marin, Patru, Manolea, et al., 2021;Suguna & Sukunya, 2019). The presence or absence of a yolk sac during the second month of human gestation was also found to be a predictor of "good or bad" pregnancy outcome in the first trimester of pregnancy (Doubilet, Phillips, Durfee, & Benson, 2022). However, whether this also results in functional changes of the yolk sac is as yet unknown. How primary damage to the yolk sac may affect the human embryo is also yet to be established.
In this review, we will discuss some of Dr Brent's research but specifically address the following topics: Development and structure of the yolk sac; the role of the yolk sac in embryonic nutrition; the role of the yolk sac in the immune and hematopoietic development and the involvement of the yolk sac in pregnancy complications and teratogenicity. As Dr. Brent conducted most of his studies in rodents, we will similarly focus our review on mice and rats.

| DEVELOPMENT AND STRUCTURE OF THE VISCERAL YOLK SAC IN RODENTS
As development begins with the embryo emerging from a few cells to the trilaminar embryo, many support cells are evolving as well: the trophectoderm surrounding the embryo and the cells of the yolk sac. With the evolution of these two support structures (chorio allantoic placenta and the yolk sac), it is noted that their functions play critical roles in mammals and other species. The focus of the work will be on the yolk sac in several species from rodent to primates. The importance at different stages of development as noted for different species will be explored (Figures 1-4).
F I G U R E 1 Carnegie Stage 7. E 7.0-7.5 Mouse embryo with surrounding membranes and ectoplacental cone. (Carnegie Collection, adapted from Ross & Boroviak, 2020). The visceral yolk sac at this stage will begin providing the nutrition for the developing embryo.
In the context of teratogenesis, the investigations have been principally on the developing embryo itself, yet the extraembryonic membranes as well as the placentae can play fundamental roles in embryonic development. For example, in the rodent, the early placenta (ectoplacental cone) is critical for the attachment of the embryo to the endometrium; however, during this early stage, the yolk sac is playing a vital role in nutrition. As both the placenta and yolk sac mature the visceral yolk sac everts itself and collects nutrients from the uterine environment and thus continues to assume a nutritional role as a transporter and metabolizer of proteins providing critical amino acids to the embryos development; (see Thowfeequ and Srinivas, 2022 for graphic descriptions of this embryonic development of the yolk sac). Interference with these yolk sac functions can compromise the development of the embryo (Brent et al., 1971;Freeman and Brown, 1980;1983;. Of special note is that the yolk sac turns itself inside out and begins to absorb nutrients, proteins from the uterine environment. Similar processes are noted in the lagomorph (rabbit). It is critical to note that the yolk sac along with its vitelline circulation provides a second supply line for the embryo/fetus in addition to the chorioplacenta and its umbilical circulation. In rodents, the yolk sac persists throughout gestation and plays a critical role in not only providing nutrients but also in the transfer of immunoglobulins to the fetus. F I G U R E 2 Schematic Representation of a cross section of the feto-placenta unit of the rat at day 11 of pregnancy. The conceptus (both embryo/fetus), extraembryonic membranes and chorio allantoic placenta are contained within the uterine wall (UW). The embryo is not depicted, but the umbilical vessels are shown that connect the chorioallantoic placenta to the embryo/fetus. The chorioallantoic placenta consists of the labyrinth and the maternalfetal junction (MFJ). The labyrinth or region for maximal maternalfetal exchange during this developmental period has both the fetal vessels (FV) within placental villi which are surrounded by the maternal blood space compared with the ectoplacental cone in Figure 1. The other extraembryonic membranes that surround the fetus are amnion (AM), visceral yolk sac (VYS), Reichert's membrane (RM) and the parietal yolk sac (PYS). Surrounding these extraembryonic membranes are the maternal decidual tissues (D). The visceral yolk sac connects with the conceptus via the vitelline vessels (VV). This is the alternate circulation for maintaining the fetus. On the fetal side of the Reichert's membrane are the parietal endodermal cells (PE) which maintain Reichert's membrane until day 17-18 of gestation, when both the Reichert's membrane and the parietal yolk sac rupture and retract to the periphery of the chorioallantoic placenta. At this time the visceral yolk sac turns inside out to allow absorption of proteins and immunogloblins to enter the fetus via the vitelline circulation. (Drawing redrawn from Miller, Koszalka, & Brent, 1976; orginally adapted from Anderson, 1959).
F I G U R E 3 Cross Section of the rat extraembryonic membranes at 14 -day old implantation site. F is a section of fetal tissue. Surrounding the fetus is the amnion (AM), visceral yolk sac (VYS), parietal yolk sac (PYS) and Reichert's membrane (RM). Note the columnar epithelium (CE) of the visceral yolk sac. The membrane is active in concentrating substances via active transport, pinocytosis and phagocytosis. The parietal yolks sac consists of Reichert's membrane, endo parietal cells and giant trophoblast cells as well as the maternal blood space. These cells synthesize the collagen necessary to produce Reichert's membrane (Magnification Â125) (from Miller et al., 1976). For a review on the development and morphology of yolk sac placenta in rodents see Jollie WP (Jollie, 1990).
In particular immunoglobulins -(IgG) are specifically transferred across to the fetus to provide important protection for the newborn against some infections. It is known that the human placenta has IgG Fc receptors on the syncytiotrophoblast of the placenta (Schneider & Miller, 2010). Of significance, the transfer of IgG occurs only across the visceral yolk sac in the rodent and rabbit (Miller et al., 1976;Ross & Boroviak, 2020). In the context of studying teratogenesis in multiple species, these relationships between the yolk sac and the chorioplacenta can play fundamentally different roles, for example, immunoglobulin and vitamin transfer (Polliotti, Panigel, Miller, 1997;Ng, Miller, Catus, 1981;Perez D'Gregorio, R, Miller, 1998).
In the human, the yolk sac also appears early with the placenta playing a critical role in attaching to the maternal endometrium. Of note is that maternal uterine circulation does not directly feed the placenta until between 8 and 10 weeks. In the human and non -human primate, the yolk sac plays vital roles early, for example, hematopoiesis; however, function decreases and the yolk sac becomes vestigial as the uterine circulation begins to play an essential role (Luckett, 1978;Ross & Boroviak, 2020).
Thus, different mechanisms of teratogenesis may develop, depending upon the critical functions played by the yolk sac at different stages of development in multiple species.
This review will not be considering the avian yolk sac; however, the chicken embryo has been a model for not only cardiac development (Hu, Ngo, & Clark, 1996) but other organ processes earlier identified in the yolk sac (Wong & Uni, 2021).

| Immunoglobulin transfer by the yolk sac
As mentioned earlier, the yolk sac is fundamental to the early immune status of rodents and lagomorphs, which is a major difference with the human. The critical factors in the transfer of IgG are the Fc binding sites on the maternal surface of the yolk sac leading to the internalization of the IgG via receptor mediated endocytosis throughout the last week of pregnancy for rodent and rabbit. Even though a similar receptor mediated endocytosis is noted for the human, the transfer in the human occurs principally after 25 weeks of gestation via the placenta since the human yolk sac is no longer present. (Schneider & Miller, 2010). Thus, mother's IgG is now present in the fetus protecting it at birth. This important role has been recently highlighted with Respiratory Syncytial Virus (RSV) vaccination of the mothers during pregnancy being protective for the newborn against RSV infection (Simoes et al., 2022).
The yolk sac in rodents has important roles as a functional placenta almost to the end of pregnancy. It is involved in embryonic nutrition, immunoglobulin transfer via receptor mediated endocytosis, hematopoiesis and F I G U R E 4 Drawing of the early mammalian placenta demonstrating critical functions for the development of the embryo and yolk sac. Note the yolk sac is the source of histotrophic nutrition early in development as is the origins of hematopoiesis and germ cells. Figure redrawn and adapted from Ross & Boroviak, 2020. establishment of the vitelline circulation. Disruption of its functions may severely hinder the development of the embryo/fetus resulting in growth restriction, malformations and/or fetal death. In human, the yolk sac plays similar roles but its functions are limited since it involutes by the end of the second month post fertilization (around 7-8 weeks).

| The role of the yolk sac in early hematopoiesis and vascular formation
The yolk sac appears to be the exclusive site of the first stage of embryonic primitive erythropoiesis in mammals (Yamane, 2018). The origin of these erythroid progenitors in the extra embryonic mesodermal cells is, most likely, from the embryonic epiblast (Yamane, 2018). The first wave of hematopoietic progenitor cells includes mainly erythroid progenitors that emerge at the stage of primitive streak formation of embryonic day 7.5 in the mouse (Palis, Malik, McGrath, & Kingsley, 2010;Yamane et al., 2001;Yamane, Washino, & Yamazaki, 2013). Primitive macrophages can be observed in the yolk sac extraembryonic mesoderm from that time (Palis, Robertson, Kennedy, Wall, & Keller, 1999). These cells form early blood islands inside the extraembryonic mesoderm of the yolk sac. The surrounding mesodermal cells differentiate to endothelial cells forming the early yolk sac vasculature and the vitelline circulation. The primitive erythroid cells start entering the mouse embryonic circulation with the onset of heart beats and cardiac contractions on day 8 post fertilization, at the stage of 4-6 somites (Ji, Phoon, Aristizabal, et al., 2003;Lucitti, Jones, Huang, et al., 2007).
These primitive erythroblasts accumulate hemoglobin, which is different from fetal hemoglobin, expressing embryo specific beta-globin. By day 9 in the mouse, with the development of the neural tube, cells expressing megakaryocyte markers also start to appear in the blood islands of the yolk sac (Tober, Koniski, McGrath, et al., 2007). During the same time, primitive macrophages also appear in the blood vessels. The first circulating blood cells inside the primitive blood vessels are the erythroidmyeloid progenitors, generally of erythroid, megakaryocyte, and macrophage lineages (Palis et al., 1999;Palis et al., 2010). The erythroid stem cells found in the murine circulation during mid-gestation are also the origin of B and T lymphocytes (Auerbach, Huang, & Lu, 1996). It should be noted that the dividing primitive erythroid-myeloid cells are found in the blood stream of mouse embryos only up to days 13-14 and from that time they are generally found only in sites of formation of blood cells (liver, bone marrow ext.), where new lineages of hematopoietic cells develop (Wittamer & Bertrand, 2020).
The notion that the yolk sac -derived hematopoietic cells are completely replaced in the circulation by hematopoietic cells originating from other sites like the liver and bone marrow is now questioned. Recently, Neo et al (Neo et al, 2021) summarized the data regarding the origin of hematopoietic stem cells in the bone marrow of adult mice. Using specific molecular markers, the investigators found that tissue -resident macrophages (TRM) originate from several sources, including the yolk sac derived hematopoietic stem cells. Such cells are found among the TRM in the alveoli and interstitial lung tissue. However, these cells seem to constitute a minority among the total TRM. Similarly, yolk sac hematopoietic cells are the origin of several TRM in other tissues including the brain, gonads and bone (osteoclasts).
Ito et al (Ito, Hikosaka-Kuniishi, Yamazaki, & Yamane, 2022) studied the origin of tissue macrophages in mice using various cell-specific antibodies. They found that the VYS' CSF1 receptor positive cells (CSF1R+) progenitor cells appear in the mouse yolk sac at embryonic stage 9 and apparently enter embryonic blood circulation and colonize the hepatic primordia and stage 10-11. They then spread to the embryonic organs to form osteoclasts and tissue macrophage cells. Interestingly, placental macrophages (fMacs) also originate from yolk sac blood islands endothelial cells (Chen, Tang, et al., 2022), similar to the origin of the placental Hofbauer cells (macrophages) in the human placental villi.
The primitive blood vessels developing in the mouse VYS form and remodel with the beginning of heart beats, as blood circulation is needed for vascular remodeling (Lucitti et al., 2007). Vessel remodeling is dependent on entrance of erythroblasts from the blood islands into the blood vessels. By lowering the hematocrit and reducing shear forces, vessels remodeling ceased and it was restored by restarting hemodynamic forces (Lucitti et al., 2007).
The sequential activation and expression of several important genes is crucial for the normal development of yolk sac hematopoiesis and vascular formation. Among them are Wnt signaling, Fibroblast Growth Factor (FGF), Bone Matrix Protein (BMP) and primitive streak genes like FOXF1 and MIXL1. Hematopoietic progenitor cells express SCL/tal-1 and GATA-1, genes that are known to be involved in the development of hematopoietic cells. These hematopoietic progenitor cells in rodents decrease in number with further embryonic development and disappear around the 20 somite stage, being replaced by erythroid progenitor cells in the yolk sac expressing c-myb gene (Palis et al., 1999). Goumans et al., (1999) demonstrated that transforming growth factor beta (TGF-β) signaling is also essential for the formation of blood vessels in the yolk sac. It regulates production and deposition of fibronectin and is involved in the organization of the primitive endothelial cells to vessels. However, TGF-β is not needed for the differentiation of the stem cells into endothelial cells (Goumans, Zwijsen, van Rooijen, et al., 1999;Palis et al., 1999). More recently, Duan et al. have demonstrated (Duan, Wang, Mitchell-Silbaugh, et al., 2019) that heat shock proteins, including heat shock protein 60 are also involved in the regulation of yolk sac erythropoiesis in mice.
Of special importance is the role of vascular endothelial growth factor A (VEGFA) in the first steps of hematopoietic and endothelial cell formation. Insufficient VEGFA activity in the yolk sac endoderm in mice resulted in severe abnormalities of embryonic vascular development (Damert, Miquerol, Gertsenstein, Risau, & Nagy, 2002).
In the human and nonhuman primates, early hematopoietic islands appear in the yolk sac in a similar pattern to the initial phases in rodents. The primary cells that develop in the blood islands give rise to erythrocytes, macrophages and megakaryocytes (Ross & Boroviak, 2020).
The yolk sac is the initial site of hematopoiesis in many animal species including mammals. Cells in the extraembryonic mesoderm of the yolk sac, of epiblastic origin, form blood islands that gradually become the source of most hematopoietic stem cells as well as the endothelial cells of vitelline blood vessels. This primary site of blood formation seems to be active until other embryonic organs (i.e., liver, bone marrow) become functional. Several genes and growth factors are responsible for these functions.

| YOLK SAC AND EMBRYONIC/ FETAL NUTRITION
The concept that the human yolk sac is an important source for early human embryonic nutrition was introduced by several investigators (Exalto, 1995;Jauniaux et al., 1994). This concept was strengthened by the finding that in human, the intervillous space does not contain blood before the 10th week of gestation (Exalto, 1995). Indeed, most studies on the nutritional role of the yolk sac were performed in rodents as the yolk sac in human was considered to be a vestigial organ as it involutes around weeks 8-10 of pregnancy. However, Cindrova-Davis et al (Cindrova-Davies et al., 2017) have shown by RNA sequencing that the human yolk sac has many similarities to the murine yolk sac, especially in relation to uptake and processing of macro and micronutrients and their transfer to the celomic fluid of the developing embryo. Hence, the data on the nutritional role of the yolk sac in rodents may be relevant to human as well; however, this role requires further investigation.
The nutritional role of the yolk sac in rodents was assessed by several investigators more than 50 years ago, especially by in vitro culture of rat yolk sacs during midgestation and fetal periods (Williams, Kidston, Beck, & Lloyd, 1975). The establishment of in vitro culture of early somite rat and mouse embryos enabled one to directly study the transfer of nutrients to the embryo by the yolk sac (Freeman & Lloyd, 1980). Active pinocytosis by the rat yolk sac at mid-gestation was apparently first shown by Felix Beck and associates in 1967 (Beck, Lloyd, & Griffiths, 1967). Horseradish peroxidase was incorporated into the visceral yolk sac epithelial cells but not in the fetal tissues, possibly demonstrating the breakdown of the peroxidase by the yolk sac. Williams et al. (Williams et al., 1975) cultured visceral yolk sacs obtained from 17.5 days old rat fetuses with the addition to the culture medium of 125 I labeled albumin and studied the rate of pinocytosis by the visceral yolk sac epithelial cells. They found a substantial uptake of the albumin into the epithelial cells. The culture medium then contained iodine labeled iodothyrosine as a measure that the yolk sac degraded the albumin, apparently demonstrating that the degradation products of albumin are transferred to the fetus.
The ability to directly study the possible degradation of proteins by the yolk sac and transfer of the degradation products to the embryo was also demonstrated by Freeman and Lloyd (Freeman & Lloyd, 1983) in cultures of 9.5-day-old rat embryos in the presence of 3 H leucine labeled hemoglobin. They demonstrated the uptake of the labeled hemoglobin in the visceral yolk sac. The hemoglobin within the yolk sac was digested releasing into the embryo 3 H -labeled amino acids that were the source for the synthesis of different proteins by embryonic tissues.
Similarly, Koszalka et al (Koszalka, Andrew, Brent, et al., 1994) studied in rats the amino acid content in the visceral yolk sac, ectoplacental cone and embryos during days 10.5-13.5 of gestation and found that inhibition of the protein/amino acid pathway can similarly affect all three structures and inhibition of pinocytosis may affect the amino acids composition in the three organs and interfere with embryonic development. It was also repeatedly shown that yolk sac dysfunction may interfere with embryonic development causing embryonic death and/or congenital malformations (Brent & Fawcett, 1998). Beckman et al. (Beckman, Lloyd, & Brent, 1997) found that endocytosed proteins were "digested" in the yolk sac lysosomes which released amino acids supplying the embryo. Visceral yolk sac pinocytosis was studied by evaluating the uptake of 14 C-sucrose during different phases of rat fetal development. Uptake of sucrose, defined by the investigators as the endocytic index, is a marker for the uptake of different nutrients by the yolk sac epithelial cells (Beckman, Lloyd, & Brent, 1998).
It can be concluded that the visceral yolk sac is the main supplier of most nutrients to the developing rodent embryo (Zohn & Sarkar, 2010). In primates, including human, the yolk sac serves for embryonic histotrophic nutrition in the first two months, only before the establishment of the intervillous maternal blood circulation (Jauniaux et al., 1994).
The yolk sac is a major route of embryonic nutrition in many animal species apparently as long as it functions. In mammals and especially rodents, macro and micronutrients are absorbed by active pinocytosis into the visceral yolk sac and then transferred to the embryo. They serve the main source of amino acids and other nutrients for the developing embryo. Nutrients are transferred to the intraembryonic celom or to the embryonic/fetal circulation via the vitelline vessels.

| Anti-embryonic and anti-yolk sac antibodies in rodents
Placental damage is known to interfere with embryonic and fetal growth and, if severe enough, may even produce intrauterine death. Whether placental dysfunction may cause embryonic malformations was an open question for many years.
Since in rodents, the yolk sac placenta has important functions throughout pregnancy, several investigators studied the possible role of the yolk sac in teratogenesis, among them -Dr. Brent and associates (Beckman et al., 1991;Brent & Fawcett, 1998).
One of the early examples in human that antibodies which cross the placenta may affect embryonic and fetal development is the "hemolytic disease of the newborn" (or Erythroblastosis fetalis) caused by Rh incompatibility. The anti Rh antibodies may induce in the fetus/newborn infant severe anemia, jaundice and sometimes intrauterine fetal death, due to severe hemolysis (Myle & Al-Khattabi, 2021). Similarly, several maternal autoimmune diseases, especially Systemic Lupus Erythematosus (SLE) and/or antiphospholipid syndrome (APLS), may produce recurrent abortions and fetal death (Fischer-Betz & Specker, 2017).
Dr. Brent and associates' first studies on the possible role of antibodies in teratogenesis were with sheep antirat kidney antibodies injected into pregnant rats on day 8 of gestation, demonstrating that they are teratogenic in rats (Bragonier, Frank, & Brent, 1970). They found that the antibodies were localized in the embryo demonstrating IgG transfer through the yolk sac. They then proved that different teratogenic antibodies ("teratogenic antisera") were also localized on the yolk sac and interfered with yolk sac pinocytosis. Interestingly, nonteratogenic antisera did not interfere with yolk sac endocytosis (Lerman et al., 1986).
Thereafter, Dr. Brent and his group used anti-yolk sac antibodies which also induced embryonic anomalies. They found that anti-yolk sac antibodies were more potent teratogens and generally did not pass to the embryo, demonstrating the importance of the yolk sac in embryonic nutrition. These observations then opened the way to study (in vitro) the mechanisms involved in embryonic nutrition via the yolk sac and demonstrate the transfer of amino acids after the degradation of proteins by the yolk sac (Beckman et al., 1997;Brent & Fawcett, 1998). While teratogenic anti-yolk sac antibodies caused definitive embryonic anomalies, the ultrastructural and functional yolk sac damage induced by these antibodies slowly disappeared 48-72 hr after exposure to the antibodies, both in vivo and in vitro (Beckman et al., 1991).
Freeman and Brown (Freeman & Brown, 1994) also used rabbit anti-yolk sac antiserum injected to rats on day 9.5 of pregnancy. The rat embryos cultured on serum from the treated rats were growth retarded and exhibited absence of optic vesicles and abnormal forebrains. In addition, yolk sacs incubated in serum from treated rats had 50% decreased pinocytosis.

| Human maternal diseases causing placental and yolk sac damage in rat embryos
Following Dr. Brent's studies in rats which demonstrated the pathological effects of various antibodies on the yolk sac and embryonic development, we investigated the possible effects of human serum and autoimmune antibodies on the developing rat embryo and yolk sac This is especially evident in autoimmune diseases, particularly Systemic Lupus Erythematosus (SLE) and antiphospholipid syndrome (APLS). Maternal SLE and APLS are known to increase spontaneous abortions, prematurity, early fetal death and stillbirths, while maternal diabetes increases the rate of congenital malformations and may induce a variety of pregnancy complications. In most cases the pathological changes were ascertained in the chorioallantoic placenta (Ornoy et al., 2003;Ornoy, Chen, Silver, & Miller, 2004).
Recently Marin et al. (2021) demonstrated that increased yolk sac diameter is associated with pregnancy loss in human. To our knowledge, no human studies on the yolk sac of women with autoimmune diseases (SLE/APLS) have been published.

| Antiphospholipid antibodies as teratogenic agents affecting rat yolk sac and embryo
A variety of IgG antibodies, especially antiphospholipid antibodies in SLE and/or APLS may negatively impact the pregnancy (Fischer-Betz & Specker, 2017;Tincani, Bazzani, Zingarelli, & Lojacono, 2008);. This is generally related to severe placental damage induced by these autoimmune antibodies (Ornoy et al., 2003;Ornoy et al., 2004). Circulating maternal antibodies enter the intervillous space, attach to the trophoblastic cells and interfere with placental function by the induction of intervillous thrombosis and/or by reducing the hormone secretions of the trophoblastic cells as observed in placental villous culture (Schwartz et al., 2007). These antibodies do not appear to cause embryonic malformations but IgG antibodies in patients with SLE such as anticardiolipin, anti -Ro and anti-La may induce fetal cardiac arrhythmia, cardiac block, or temporary clinical features of SLE observed in the newborn infant.
We conducted several studies using early somite rat embryo cultures where we examined the effects of serum from women with SLE/APLS or used autoimmune antibodies to evaluate embryonic development, yolk sac structure and function. Culture of 11.5 day -old rat embryos for 24-48 hr in a medium containing serum from women with SLE or APLS induced a high rate of embryonic malformation and death and, in addition, also ultrastructural and functional damage to the yolk sac in comparison to embryos cultured on serum from healthy pregnant women. The yolk sac damage was manifested by large intracytoplasmic inclusions in the endodermal epithelial cells of the visceral yolk sac, denuded cells, loss of microvilli and decreased endocytosis (Ornoy et al., 2003;Ornoy, Yacobi, Avraham, & Blumenfeld, 1998). These pathological changes were attributed to the damaging effects of different autoantibodies. Indeed, similar yolk sac damage, including reduced embryonic and yolk sac growth, was observed when the embryos and their yolk sacs were cultured in rat serum but in the presence of different antibodies obtained from women with SLE/APLS and recurrent pregnancy loss (Matalon et al., 2002;Ornoy et al., 2003). Anti-cardiolopin and anti-DNA antibodies induced a more prominent growth retardation and yolk sac damage in comparison to anti-phosphatidyl serine and anti-laminin. It is interesting to note that a high rate of embryonic malformations and ultrastructural yolk sac damage (low number of microvilli and a high number of intracytoplasmic yolk sac endodermal cells) was also observed when early somite rat embryos were cultured in serum from women with recurrent abortions but not diagnosed with SLE or APLS (Abir, Ornoy, Ben Hur, et al., 1994).

| Maternal diabetes, hyperglycemia, and yolk sac damage
Pathological changes in the placenta have been described in maternal pregestational diabetes for many years, especially in women with diabetic complications like nephropathy or with cardiovascular complications (Vambergue & Fajardy, 2011). Changes in yolk sac diameter were described in pregnancies of women with type 1 diabetes. Ivanisevic et al., (Ivanisevic, Djelmis, Jalsovec, & Bljajic, 2006) described an increase in the diameter of the yolk sac in pregnant women with type 1 diabetes and Cosmi et al., (Cosmi et al., 2005) found delayed decrease in its diameter during involution by the end of the second month post fertilization However, Papaioannou et al., did not find such differences in women with pregestational or gestational diabetes compared to nondiabetic pregnant women (Papaioannou, Syngelaki, Nerea Maiz, et al., 2012).
These findings led several investigators to suggest that the yolk sac damage in maternal diabetes may play an important role in diabetic embryopathy (Dong et al., 2016;Reece, Pinter, Homko, Wu, & Naftolin, 1994). Moreover, abnormal size and shape of the yolk sac, or its absence, were also found as predictors of general poor pregnancy outcome in human (Suguna & Sukanya, 2019).
We cultured early somite rat embryos in serum obtained from women with pregestational (PGD) or gestational (GD) diabetes compared to serum from healthy pregnant women (Ornoy, Zaken, Abir, Yaffe, & Raz, 1995) and observed a high rate of embryonic malformations. There was a reduction of embryonic and yolk sac size with decreased protein content. The damage was more severe when using serum from women with PGD compared to serum from women with GD. Studies conducted in vivo on diabetic rats and in vitro using rat embryo cultures revealed severe damage to the visceral yolk sac endodermal cells that included degenerated and denuded endodermal cells, reduced microvilli and an increased number of intracytoplasmic inclusions (Ornoy, Zusman, Cohen, & Shafrir, 1986;Zusman, Yaffe, & Ornoy, 1987).
High glucose levels are known to increase embryonic and yolk sac oxidative stress as observed both in vivo and in vitro (Ryu, Kohen, Samuni, & Ornoy, 2007). The addition of nitroxide-an effective antioxidant-to the diabetic culture medium prevented the embryonic and yolk sac damage induced by the diabetic rat serum and also normalized yolk sac endocytosis as observed by the normalization of glucose uptake in the yolk sac (Ryu et al., 2007). Thus, enhanced hyperglycemia-induced oxidative stress may be responsible for the yolk sac damage and potent antioxidants like nitroxides reduce that damage. Moreover, the yolk sac damage may aggravate or at least contribute to the embryonic damage induced by hyperglycemia.
Similar to our studies, Pinter et al.,  found that high glucose levels in cultured rat embryos induced VYS damage manifested by reduced endoplasmic reticulum and sparse mitochondria in the endodermal yolk sac epithelial cells. The yolk sac damage was prevented by the addition of arachidonic acid to the culture medium . The investigators concluded that the yolk sac damage hinders yolk sac function which subsequently causes embryonic malformations (diabetic-induced embryopathy).
High glucose also reduced vascular endothelial growth factor A (VEGF-A) expression and VEGF receptor activation in the yolk sac, possibly explaining the mechanism of the induced yolk sac vasculopathy (Pinter, Haigh, Nagy, & Madri, 2001). Reece et al., (1994) also found that high-glucose levels reduced the uptake of horseradish peroxidase in the visceral yolk sac of cultured rat embryos, further demonstrating the yolk sac functional damage.
Hunter and Sadler (Hunter & Sadler, 1992) cultured head fold stage mouse embryos in a medium containing different high concentrations of glucose and found neural tube closure defects in a dose response manner. The uptake of 35 S-methionine and 35 S-cysteine-labeled hemoglobin by the yolk sac following short periods of culture was normal. However, the accumulation of 35 S in the embryo was reduced. Exposure of the yolk sac to hyperglycemia for longer times resulted in reduced yolk sac uptake.
Nath et al., (Nath et al., 2004) found in mice that diabetes interferes with yolk sac vasculogenesis inducing vitelline vasculopathy. They also found that high glucose increased the levels of reactive oxygen species (ROS) increasing yolk sac and embryonic oxidative stress. It also increased NO production inducing vitelline vasculopathy.
Moreover, administration of an NO donor prevented the hyperglycemia-induced yok sac vasculopathy. Dong et al., (2016) in a literature survey regarding yolk sac damage following maternal diabetes in human as well as in animal models, hypothesized that hyperglycemia may induce yolk sac damage to a significant degree which will lead to vitelline vasculopathy. This is backed by a significant number of studies in human and in rodents. The molecular mechanisms that may be responsible are a reduction in VEGF-A expression, reduced VEGF receptor activation leading to altered signaling in the VEGF-A pathway resulting in vitelline vasculopathy.
Interference with the yolk sac function may reflect directly on embryonic growth and development, inducing congenital malformations or causing, in extreme damage, embryonic and fetal death. In rodents many agents were found to damage the yolk sac (i.e., anti-yolk sac antibodies or toxic substances interfering with yolk sac pinocytosis) subsequently affecting the embryo/fetus. Often, the damage to the yolk sac is transient. In humans, decreased yolk sac diameter was demonstrated in diabetic pregnancies and increased diameter is associated with pregnancy loss. In addition, in vivo and in vitro studies on the rat VYS in diabetes produced severe damage to the visceral yolk sac epithelium. Similar yolk sac pathology was found when rat embryos were cultured in serum from women with SLE, APLS or diabetes or after the addition of autoimmune antibodies to the culture medium.

| CONCLUSION
For preclinical Teratology studies, the rat has been an animal model for examining exposures to therapeutic and environmental agents. Besides the differences in gestational age and development at birth, the yolk sac is a major difference between human and rodent. These differences are examined in detail for anatomy and function during development related to hematopoiesis, nutrition, metabolic, immunologic, and specific toxic responses. The literature related to the human yolk sac in normal and abnormal pregnancies is mainly related to its size. Hence, there is a need for functional and molecular studies related to the role of the yolk sac in normal and abnormal pregnancies. Such studies may be done in vitro using yolk sac tissue from early human spontaneous abortions.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.