Introduction

During early mammalian development, the first lineage distinction occurs at the blastocyst stage, with the separation of the outer trophectoderm (TE) from the inner cell mass (ICM) (1). The TE goes on to form the extra-embryonic ectoderm and ectoplacental cone after implantation; these trophoblast tissues form the majority of the fetal-derived components of the mature placenta. The ICM gives rise to the fetus itself, as well as some other extra-embryonic cell types. The trophoblast supports the growth and development of the fetus in the uterine environment, by facilitating nutrient, gas, and waste exchange in the placenta, ensuring immune protection of the fetus, and producing a variety of hormones and cytokines that influence both maternal and fetal systems (2). In addition, the early postimplantation trophoblast, prior to mature placental formation, is a key player in the complex signaling interactions required to establish the basic body axes in the developing embryo itself (3). Given its complex roles, it is not surprising that many genetic pathways impinge on trophoblast development and function. However, many of these pathways are also involved in the development of the embryo itself; genetic knockout studies often fail to separate the different roles (4). Indeed, in some well-known cases, genes have been incorrectly proposed to be necessary for the development of specific embryonic tissues, based on phenotypes that later turned out to be entirely secondary to defects in the placenta (5,6).

Embryonic stem (ES) cell-derived embryos, made by aggregating ES cells with tetraploid embryos, can be used to separate the action of mutations in embryonic versus extraembryonic lineages (7). However, this system fails to separate functions in the trophoblast versus the extraembryonic endoderm and does not address very early lineage specific effects. Trophoblast stem (TS) cell chimeras could in theory be used to manipulate gene expression specifically in the trophoblast, but it is difficult to obtain consistently high levels of chimerism within the trophoblast to make this a viable option (8). Gain of function or rescue of gene function analysis could be achieved by standard transgenesis, using trophoblast-specific promoter-driven transgenes. Loss of function could be achieved using the same promoters to drive Cre or Flp recombinases for conditional gene targeting (9). However, there are limited numbers of well-defined trophoblast-specific promoters and none that is expressed throughout the trophoblast lineage at all stages of development. These approaches are also tedious and slow, requiring multiple rounds of mouse breeding. We aimed to develop a simple technique for rapid and facile gene function analysis in the trophoblast, to complement the use of ES-cell derived mice.

While non-lentiviral retroviruses have long been known to infect embryos at the preimplantation stage (10), lentivirus vectors have been shown to infect early mouse embryos more efficiently; up to 80% of embryos were reported to show widespread transgene expression at later stages after infection of one-cell, two-cell, or morulae (11). The same vectors can infect ES cells and seem to be less prone to silencing than other retrovirus vectors (12). We reasoned that infecting not morulae but blastocysts could prove an effective way of achieving trophoblast-restricted gene expression from the blastocyst stage onwards. Passage of large molecules into the ICM and the blastocoelic cavity is known to be prevented by the epithelial tight junctions of the TE. This suggested that viral particles would also be excluded, thus allowing infection of only the outside TE cells. There has been one report that describes the use of a lentiviral vector to infect the TE of rhesus monkey (Macaca mulatta) at the blastocyst stage (13). The blastocysts were infected by microinjection into the blastocoel, which led to the infection of the both the TE and the primitive endoderm derived tissues. We present here a methodology for the infection of only the TE layer by culturing the blastocysts in media with lentiviral vectors.

Materials and Methods

Generation of Virus

Plasmids for generation of the FUGW UbiC enhanced green fluorescent protein (EGFP) (11) and FG12 U6 RNA interference (RNAi) LacZ lentivirus (14) were kindly provided by Drs. Carlos Lois (Division of Biology, California Institute of Technology, Pasadena, CA) and Xiao-Feng Qin (University of Texas M.D. Anderson Cancer Center, Houston, TX), respectively. The lentivirus expressing Cre was generated by cloning the nuclear localized improved Cre recombinase (iCre) gene (15) into the EF.v.CMV. GFP vector (16) (accession no. JHU-55; ATCC, Manassas, VA, USA) at the EcoRV restriction site. Virus was grown as published (17), and viral titer was determined as published (18).