Article
Extended Data Fig. 1 | Establishing a 3D blastocyst-culture system. This
figure is related to Fig. 1. a–f, Modification of human embryo culture medium.
To test human embryo development, human embryos at 5–6 d.p.f. were
cultured on low attachment plate without any 3D extracellular matrix up to
14 d.p.f. By sequentially culturing in IVC1 (6–8 d.p.f.) and IVC2 (8–14 d.p.f.)
media, 6.67% (2 out of 30 embryos) of embryos could survive until 14 d.p.f.
a, Schematics of improving culture medium. Sodium lactate, sodium pyruvate
and ROCK inhibitor (Y27632) were added to the IVC1 and IVC2 media, referred
to as mIVC1 (6–8 d.p.f.) and mIVC2 (8–14 d.p.f.), respectively. Culture in mIVC1
and mIVC2 enabled 25% (4 out of 16 embryos) of human blastocysts to develop
up to 14 d.p.f. (c). b, Representative developing embryos based on
morphological observation. n = 16 independent embryos from three
independent experiments. c, Representative embryos with abnormal
development. Abnormal embryos displayed growth arrest or had obvious cell
death or fragmentations. n = 13 independent embryos from three independent
experiments. d, Representative staining of abnormal embryos with CK7,
GATA6 and OCT4 at 14 d.p.f. In all six examined embryos, consistent data were
obtained. e, Staining of developing embryos (2 out of 3 embryos) with CK7,
GATA6 and OCT4 at 14 d.p.f. f, Quantification of developmental rates of human
embryos cultured in control medium (IVC1 and IVC2) and modified medium
(mIVC1 and mVIC2). n = 30 and 16 biologically independent embryos,
respectively. Developmental rates of human embryos at 11 and 13 d.p.f. were
based on the two following requirements by morphology: obvious expansion
over culture and absence of obvious cell death mass and fragmented
phenotypes. At 14 d.p.f., we determined the embryo development ratio by
staining CK7 (TrB), GATA6 (PrE,) and OCT4 (EPI). g–q, Representative human
embryo development after culture under four different 3D conditions over
development. The limited number of embryos only enabled us to compare
embryo development under four conditions. As the implantation time window
is 8–10 d.p.f., we embedded embryos with Matrigel at 9 d.p.f. Embryo
development was verified on basis of morphological observation and staining
of specific markers for OCT4, GATA6 and CK7. g–j, Top, schematics of in vitro
3D culture of human blastocysts under different culture conditions. Middle,
representative images of human embryos during development. Bottom,
representative stained images of cultured human embryos under different 3D
conditions at 14 d.p.f. g, Human embryos at 5–6 d.p.f. were cultured on low
attachment plate without Matrigel (W/O Matr) up to 14 d.p.f. The outermost
TrBs showed signs of apoptosis (blue arrowheads), as determined by
morphological observations and CK7 staining, which suggests that the
condition was unsuitable for TrB development and survival required for
attachment. In total, 2 of 20 embryos (three independent experiments)
survived up to 14 d.p.f. and displayed normal EPI, PrE and TrB development.
h, Human embryos were embedded in 25% Matrigel at 9 d.p.f. for continuous
culture up to 14 d.p.f. n = 25 embryos from three independent experiments. The
invasion and outgrowth of TrBs were observed and embryos became f lat, which
suggests a higher concentration of Matrigel is advantageous to differentiation
and development of TrBs, as confirmed by CK7 staining. i, Human embryos
were embedded in 10% Matrigel on the new well, which was pre-coated with
100% Matrigel (30 μl) (Matr+10%Matr), at 9 d.p.f. for continuous culture up to
14 d.p.f. Although embryos displayed considerable expansion over culture,
staining with lineage markers showed that EPIs in most embryos were lost over
development. Only 1 of 33 embryos from three independent experiments grew
to 14 d.p.f and was accompanied by EPI, PrE and TrB development. The negative
outcome may indicate that high concentrations of Matrigel can inhibit EPI
development. In h and i, red arrowheads indicate TrBs invading into Matrigel.
j, Human embryos were embedded in 10% Matrigel at 9 d.p.f. for continuous
culture up to 14 d.p.f. Compared with the 25% Matrigel and Matr+10% Matr
conditions, human embryos in 10% Matrigel increased in size at the thickness
(Z axis) and showed better 3D spatial structures. In total, 4 of 17 embryos from
three independent experiments grew to 14 d.p.f. with normal development.
k, Quantification of the mean diameter of human embryos cultured under
different 3D conditions during development by analysis of 5–14 embryos from
three independent experiments. Data are mean ± s.d. l, Quantification of
human developmental embryos during culture in different 3D conditions. Data
were based on morphological observations only. Human developing embryos
met the two following requirements: obvious expansion over culture; absence
of obvious cell death mass or fragmented phenotypes. m, Quantification of
developing embryos in different 3D conditions. n = 20 (W/O Matr), 25 (25%
Matr), 33 (Matr+10% Matr) and 17 (10% Matr) blastocysts. Developing embryos
met the following requirements: obvious expansion over culture; absence of
obvious cell death mass or fragmented phenotypes; and development of EPIs,
PrEs and TrBs and formation of amnion identified by OCT4, GATA6 and CK7
staining. Although embryos in the 25% Matr and Matr+10% Matr culture
conditions have normal morphologies, some embryos lacked OCT4+ EPIs or
G ATA6+ PrEs and gave rise to a high proportion of TrBs, which suggest that high
concentrations of Matrigel could inhibit EPI development and promote TrB
proliferation. n–q, Three-dimensional construction of human embryos
cultured in 10% Matrigel (3 out of 3 embryos from three independent
experiments). n, A representative 3D reconstruction of EPIs. Inset shows OCT4
staining of one section from the same embryo. o, A representative 3D
reconstruction of SYS and amnion including an embryonic disc and an amniotic
cavity (see Supplementary Video 1). p, Three-dimensional reconstruction of
TrBs (see Supplementary Video 2). q, Three-dimensional magnification of TrBs
close to the amnion side. Scale bars, 100 μm (phase-contrast) or 50 μm
(staining).