Background
Rodents are useful models for human reproduction due to the ready availability of laboratory animals [
1] and their closeness to the primate lineage [
2‐
4]. Although four suborders are recognized, most species used in research are myomorph rodents [
1]. A notable exception is the guinea-pig, which is a hystricognath rodent from the suborder Hystricomorpha [
5]. The hystricognath rodents have adopted a reproductive strategy characterized by a relatively long gestation, small litter size and the delivery of well-developed (precocial) young [
6]. This is in many respects similar to reproduction in higher primates [
7]. For this reason among others [
1,
8], they offer more satisfactory models for human pregnancy than rodents that have short pregnancies and deliver large litters of poorly developed (altricial) young. As an example, events occurring during later stages of pregnancy in humans must be studied postnatally in rats and mice, introducing a wealth of confounding factors. There are several similarities in placentation between hystricognaths and higher primates including a single layer of syncytiotrophoblast in contact with the maternal blood space (i.e. haemomonochorial) as opposed to three trophoblast layers (i.e. haemotrichorial) in myomorph rodents. There are as well similar patterns of trophoblast invasion and placental growth [
1,
9‐
13].
Current concepts of palaeogeography favour an African origin for hystricognaths with dispersal to South America by a trans-Atlantic route in the Eocene or Oligocene [
14]. The subsequent radiation resulted in the wide range of forms found in South America today [
15,
16]. The semi-aquatic capybara (
Hydrochaeris hydrochaeris) is by far the largest extant species of rodent. Like other hystricognaths, it delivers precocial neonates after a relatively long gestation period [
17,
18].
Although the guinea pig is an attractive model for human pregnancy, the question arises whether it is possible to compare such a small animal with the condition in humans. To better understand this we have studied placental development in the capybara, which more closely approximates human dimensions with a maternal body mass around 50 kg, a delivery weight of around 1 kg and a gestation period of around 150 days [
18]. The main aim of the study is to substantiate if the principle processes of placentation depend on body size or not. Special attention was paid to the following questions: How is the lobulated arrangement of the placenta developed in the capybara? Previous studies had shown only the architecture of the term placenta [
19‐
21]. Does the labyrinth continue to grow in the course of gestation in the same way as in smaller hystricognaths? How do the ontogenetic differentiation of the subplacenta and the associated pattern of trophoblast invasion occur? These are both specialized features of hystricognath placentation. Finally, what is the significance of these findings on placental differentiation in the capybara for the choice of smaller species as models for human placentation?
Methods
Tissue collection and fixation
The observations are based on material collected from six animals at various stages of pregnancy (Table
1). Relevant placental characteristics of the capybara and related hystricognath species investigated so far are summed up in Tables
2 and
3[
6,
9‐
13,
19‐
50].
Table 1
Fetal and placental size at the four stages of gestation studied
1.2 (n = 1) | 1.5 × 1.3 | Late limb bud stage | 1 |
5.5–5.8 (n = 2) | 3.5 × 2.5 | 70 | 2 |
8.0–13.0 (n = 3) | 6.7 × 4.0 | 90 | 4 |
Table 2
List of major placental characters and conditions in Hystricognathi
1 | Organization of main placenta | Moderate lobulation (1). Complex lobulation (2). |
2 | Capsule around placenta | Absent (1). Present until mid gestation (2). |
3 | Labyrinth I | Radial arrangement of the exchange areas around the maternal blood lacunas (1). |
4 | Labyrinth II | Countercurrent arrangement of the fetal and maternal blood flows (1). |
5 | Interhaemal barrier | Haemochorial placenta with cellular and syncytial trophoblast in early ontogeny and mostly syncytial barrier later on (1). |
6 | Interlobium | Substantial areas of interlobium associated with labyrinth (1). |
7 | Placental growth | Essentially the result of specialised growing zones at the outer margin, associated with internally directed projections on fetal mesenchyme (1). |
8 | Additional proliferation | Insignificant proliferation activity in trophoblast in the labyrinth (1). Remarkable proliferation in this trophoblast in early and mid gestation (2). |
9 | Presence of subplacenta | Distinct and specialised area, consisting of layers of cellular and syncytial trophoblast on fetal mesenchyme, occurring from early ontogeny to near term or term (1). |
10 | Blood supply of subplacenta | Associated with the maternal circulation in early ontogeny and supplemented by the fetal system later, without overlap between the two systems (1). Same composition, but some overlap between the systems during mid gestation (2). |
11 | Extraplacental trophoblast | Both cellular trophoblast and syncytial streamers in the decidua between the subplacenta and the materal arterial system, probably responsible for the replacement of the maternal arterial endothelium (1). |
12 | Visceral yolk sac | Inverted yolk sac with conspicuous villous areas (1). |
13 | Fibrovascular ring | Specialised arterial and capillary system within the yolk sac near the attachment to the main placenta (1). |
14 | Parietal yolk sac | Present, usually multi-layered from mid gestation on (1). |
Table 3
Distribution of placental characteristics in hystricognath rodents
Capybara (Hydrochaeris) | 2 | 2 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 2 | 1 | 1 | 1 | 1 | |
Guinea pig (Cavia) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 1 | 1 | |
Prea (Galea) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 2 | 1 | 1 | 1 | 1 | |
Rock cavy (Kerodon) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 1 | 1 | |
Paca (Cuniculus) | 2 | 2 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 1 | 1 | |
Agouti (Dasyprocta) | 2 | 2 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 1 | 1 | |
Chinchilla (Chinchilla) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | ? | 1 | 1 | 1 | 1 | 1 | 1 | |
Degu (Octodon) | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | |
Nutria (Myocastor) | 2 | 2 | 1 | 1 | ? | 1 | 1 | ? | 1 | 1 | 1 | 1 | 1 | 1 | |
Canadian porcupine (Erethizon) | 2 | 1 | 1 | 1 | 1 | 1 | ? | ? | 1 | 1 | 1 | 1 | 1 | 1 | |
Dassie rat (Petromus) | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
Cane rat (Thryonomys) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | ? | 1 | 1 | ? | 1 | 1 | 1 | |
African porcupine (Hystrix) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | ? | 1 | 1 | 1 | 1 | 1 | 1 | |
Mole rat (Bathyergus) | 2 | 1 | 1 | 1 | 1 | 1 | 1 | ? | 1 | 1 | 1 | 1 | 1 | 1 | |
Capybara material was collected at hysterectomy from animals bred at the Centre for Experimental Breeding of Capybaras, Paulista State University, Araçatuba, São Paulo, as authorized by the Brazilian Institute of the Environment and Renewable Natural Resources (IBAMA). Additional material (n = 3) was obtained at an IBAMA licensed slaughterhouse (Panamby-Porã, Miracatu, S.P.). The experimental protocol was approved by the Bioethics Committee of the School of Veterinary Medicine, University of Sao Paulo.
Bilateral hysterectomy was performed in 3 capybaras. The animals were premedicated with acepromazine (Univet, São Paulo, S.P., Brazil; 0.1–1.0 mg/kg I.M.). Anaesthesia was induced with xylazine (Dorcipec®, Vallée S.A., Montes Claros, M.G., Brazil; 0.5–1.0 mg/kg) and ketamine (Cristália, Itapira, S.P., Brazil; 5–10 mg/kg I.M.) and continued with halothane (Hoechst, Frankfurt, Germany; 1 per cent) or enflurane (Etrane®, Abbott, São Paulo, S.P., Brazil) in oxygen. Postoperative treatment included antibiotic coverage with benzyl penicillin and streptomycin (Pentabiotico®, Fort Dodge, Campinas, S.P., Brazil; 8000–24 000 IU/kg I.M.) and analgesia as required with flunixin meglumine (Banamine®, Schering-Plough, Rio de Janeiro, R.J., Brazil).
In one placenta from mid gestation the maternal and fetal vessels were injected with coloured latex (uterine artery white, uterine vein blue, umbilical artery yellow and umbilical vein red) in order to show the vessel distribution.
Tissues collected for histology and immunohistochemistry were immersion fixed in 10 per cent formalin in 0.1 M phosphate buffer, pH 7.4, for 24–48 h. After fixation they were submitted to dehydration and embedded in paraplast. Tissues for transmission electron microscopy were fixed in 2.5% glutaraldehyde or 2% paraformaldehyde/2.5% glutaraldehyde for 24 h and embedded in araldite as described below or in Spurr?s resin.
Histology and immunohistochemistry
The blocks were sectioned at 5 μm using an automatic microtome (Leica RM2155, Germany). Sections were stained by standard procedures with haematoxylin and eosin, Masson's trichrome and the periodic acid-Schiff reaction (PAS).
Following an approach established by Carter at al. [
51], immunohistochemistry was performed for cytokeratin to identify epithelial cells and trophoblasts; vimentin to identify mesenchymal cells and stromal decidua; and α-smooth muscle actin to identify vessel walls. As a proliferation marker we used a mouse monoclonal antibody to human proliferating cell nuclear antigen (PCNA).
Sections were dewaxed then rehydrated in an ethanol series and in the course of this they were submitted to endogenous peroxidase blockage in 3% hydrogen peroxide (v/v) in ethanol for 20 minutes. They were then placed in 0.1 M citrate buffer, pH 6.0, and submitted to microwave irradiation at 700 MHz for fifteen minutes. The sections were equilibrated in 0.1 M phosphate-buffered saline (PBS), pH 7.4, and non-specific binding was blocked using Dako Protein Block (DakoCytomation, Carpinteria, California, USA) for 20 minutes.
Tissues were incubated with primary antibodies overnight at 4°C in a humid chamber. Cytokeratin was detected by a rabbit polyclonal antibody (1:500; PU071-UP, Biogenex, San Ramon, California, U.S.A.). Mouse monoclonal anti-human primary antibodies were used to detect vimentin (1:200; V9, sc-6260, Santa Cruz Biotechnology, Santa Cruz, California, USA), α-smooth muscle actin (1:300; Clone 1A4, DakoCytomation, Carpinteria, California, USA), and PCNA (1:100; PC10, sc-56, Santa Cruz Biotechnology, Santa Cruz, California, USA). The slices were then rinsed in PBS and incubated with the biotinylated secondary antibody for 45 minutes, followed by streptavidin-HRP for 45 minutes (LSAB®+ System-HRP, DakoCytomation, Carpinteria, California, USA). After rinsing in PBS, the binding was visualized using aminoethyl carbazole (AEC Substrate Kit, Zymed Laboratories, South San Francisico, California, USA) or diaminobenzidine (DAB) as the chromagen. The sections were counterstained with haematoxylin and mounted in Faramount® (DakoCytomation, Carpinteria, California, USA) or Permount® (Fisher Scientific, Fair Lawn, New Jersey, USA). Negative controls were performed using PBS instead of primary antibody solution.
Transmission electron microscopy
Post fixation was in 2% phosphate-buffered osmium tetroxide, pH 7.4, for 2 h. Tissues were then washed in PBS (3 × 10 min) and immersed in a saturated uranyl acetate solution for 1 h. After washing in distilled water (3 × 10 min), they were dehydrated in alcohol and immersed in propylene oxide for 15 min. They were then immersed in a 2:1 mixture of propylene oxide and araldite (Polysciences Inc., Warrington, Pennsylvania, USA) for 1 h, in a 1:1 mixture for 30 min, in a 1:2 mixture for 2.5 h and in araldite for 3 h. Finally, they were embedded in araldite and baked in an oven at 70°C for 2–3 days to complete polymerization.
Semithin sections were cut at 1 μm on an automatic ultramicrotome (Ultracut R, Leica Microsystems, Nussloch, Germany) and stained with a 1% aqueous solution of toluidine blue to identify areas of interest. Ultrathin sections 70 nm thick were collected on copper mesh and contrasted with 2% uranyl acetate for 7–10 min and with 0.5% lead citrate for 7–10 min. Finally, the sections were studied in a transmission electron microscope (Morgagni 268D, FEI Company, Eindhoven, the Netherlands). Images were captured with a MegaView III camera linked to an image analysis system (Soft Imaging System, Münster, Germany).
Discussion
Among rodents, and especially within the subgroup Hystricognathi, there is an enormous variation in size. With a body weight of around 50 kg the capybara is by far the largest living rodent. Thus, the question arises of whether placental development in the capybara follows the same course as in its much smaller relatives. The capybara is the only rodent species that approximates human dimensions in body mass of mother and offspring. Our findings on placental development in the capybara are therefore a useful test of the presumed suitability of hystricognath rodents, particularly the guinea pig, as animal models for human placentation.
Firstly, how is a fetus of this size supported and how does the associated placental architecture develop? Even in rodents that deliver smaller and less well developed young, placental gas exchange is optimized by countercurrent arrangement of the maternal and fetal blood vessels. As shown by Mossman [
52], no further improvement in efficiency can be gained by increasing the length of the capillaries. The solution adopted by the hystricognath rodents was to fold the labyrinth, thus increasing the exchange area while keeping the placenta quite compact. The lobulated appearance of the placenta in cross section is a result of this and has been considered a defining feature of the hystricognath placenta [
9,
10,
21‐
23]. In the capybara, first steps towards this arrangement are apparent at an ontogenetic stage of around 70 days. Full establishment of the highly complex, lobulated appearance is achieved by 90 days, and is then equivalent to the condition of the term placenta [
19‐
21]. As in the mature placenta, by mid gestation the labyrinth shows a fully organized counter current arrangement of maternal blood channels, lined by trophoblast, and fetal capillaries. This is associated with the presence of large arterial blood channels at the centre of each lobe and interlobular areas to collect the maternal blood from the lobes. Thus, establishment of the placental architecture of the capybara is similar to that described for other hystricognath species [[
6,
9,
10,
19‐
51]; see Tables
2 and
3]. In the capybara material investigated, ranging from a limb bud stage to a mid gestation stage of 90 days, placental growth is predominantly the result of proliferation of cellular trophoblast situated in nests at the fetal side of the placenta and along internally directed projections on fetal mesenchyme. Additional proliferation has been demonstrated for cellular trophoblast inside the labyrinth. This pattern is present in later stages, too, and continues to near term (data not shown). Thus, even though placental dimensions are much larger in the capybara, the essential growth processes are equal to those in other hystricognaths, especially those with a highly lobulated placental architecture [[
11,
13], Tables
2 and
3].
Secondly, a unique feature of the placenta of hystricognath rodents is the subplacenta. Its purpose is not fully understood but one function is to act as a source for the trophoblast that invades and transforms the maternal arteries [
10,
12]. This process is important in ensuring an adequate blood supply to the placenta and is analogous to the transformation of the spiral arteries in the human placenta [
12]. As in the early development of other hystricognaths [[
6,
9,
10,
12,
19‐
29,
31‐
44,
46‐
50], Tables
2 and
3], a subplacenta associated with the maternal vasculature and characterised by layers of syncytial and highly proliferative cellular trophoblast, can be confirmed at an early stage in the capybara. At mid gestation differentiation is typical of that in other hystricognaths, but there is an overlap of fetal arteries and maternal blood lacunae that could be confirmed by injection of the vessels with latex. In contrast, the subplacenta is supplied only by fetal vessels at term. The overlap between the two systems in mid gestation is a rare feature among hystricognaths (Tables
2 and
3), known only for the degu [
35] and the prea [
31], whereas the ancient condition of hystricognaths does not include an overlap or fetomaternal exchange inside the subplacenta [
35]. All stages of placentation in the capybara that we investigated had extraplacental trophoblast, i.e. large cellular trophoblast cells and syncytial streamers derived from the subplacenta, that was responsible for the destruction and replacement of the maternal vessel walls deep within the decidua. This likewise represents a typical hystricognath feature [
10,
12,
19,
31,
35‐
37]. Thus apart from the difference in placental diameter, both the differentiation of the subplacenta and its role for trophoblast invasion is similar within hystricognaths.
Based on this and previous studies we are able to list 14 characteristics of placentation that hold across the 11 families so far studied, including African representatives such as the cane rat, African porcupine and mole rat (Table
2). As shown in Table
3, just four of these show sufficient variation between species to justify definition of two character states. This indicates an extraordinarily stable pattern that clearly was evolved before dispersal of the group to South America [
14].
Conclusion
In summary, our findings on the capybara, a rodent with around 50 kg body mass, show that all important characteristics of placental development and organization in guinea pig related rodents are similar and do not vary with body mass. Therefore it is not necessary to solve the challenges of animal husbandry that presently preclude use of the capybara as a laboratory animal. Rather, our data indicate that its smaller relatives, especially the guinea pig, are adequate models for human placentation and pregnancy.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
TCS and MAM devised the study, participated in its design and coordination and helped to write the manuscript. CK and TCS performed the major part of the histological analysis. AMC and AMM participated in the study design and analysis and wrote the manuscript. All authors read and approved the final manuscript.