The Role of the 3Rs for Understanding and Modeling the Human Placenta

Abstract

Modeling the physiology of the human placenta is still a challenge, despite the great number of scientific advancements made in the field. Animal models cannot fully replicate the structure and function of the human placenta and pose ethical and financial hurdles. In addition, increasingly stricter animal welfare legislation worldwide is incentivizing the use of 3R (reduction, refinement, replacement) practices. What efforts have been made to develop alternative models for the placenta so far? How effective are they? How can we improve them to make them more predictive of human pathophysiology? To address these questions, this review aims at presenting and discussing the current models used to study phenomena at the placenta level: in vivo, ex vivo, in vitro and in silico. We describe the main achievements and opportunities for improvement of each type of model and critically assess their individual and collective impact on the pursuit of predictive studies of the placenta in line with the 3Rs and European legislation.

Keywords: 3Rs; in silico models; in vitro models; organ on a chip; placenta; placenta on a chip.

Figure 1

Representation to show the development of advanced in vitro systems: (a) spheroids, clusters of single cells or aggregates grown without a cellular scaffold in low attachment plates; (b) trophoblast organoid grown in 3D using Matrigel; (c) a simple placenta-on-a-chip (PoC) model where two microfluidic channels are separated by a semipermeable membrane or scaffold. Created with BioRender.com.

Figure 2

Brightfield images of 3D BeWo cells grown in Growdex (Helsinki, Finland) at different stages of growth. Cells were seeded at 80,000 cells/mL. Rows of cells imaged at 6 (a), 7 (b), and 21 (c) days. Cells appear to grow in clusters and form rounded multinucleated ring structures (blue circles in (b)). Relative expression of syncytin-2 in nonsyncytialized BeWo cells grown for 7 days in 3D (Growdex) versus nonsyncytialized BeWo cells grown in 2D; ** p < 0.01 (d). Immunofluorescent images of 3D BeWo cells grown on Matrigel (e); E-Cad: E-cadherin; scale bars, 25 μm.

Figure 3

The three primary types of in silico models. Computational fluid and structure interaction models describe the architecture and hemodynamics of the placenta while physiologically based pharmacokinetic (PBPK) and quantitative structure-activity relationship (QSAR) models focus on barrier transport. Blood flow can be characterized by the Navier-Stokes and continuity equations that describe velocity, pressure gradient and blood viscosity [80]. The transfer of substances can be analyzed using a PBPK model that can generically express the change in the concentrations in the placental tissue and fetal compartments (C_p and C_f, respectively) according to the concentration in the maternal compartment (C_m), blood flow rate (Q) and the membrane diffusion (D) and partition coefficients (φ) [81]. QSAR models correlate transfer with physicochemical and structural properties using statistical methods such as multivariate analysis [82].