Completely ES cell-derived mice produced by tetraploid complementation using inner cell mass (ICM) deficient blastocysts

Abstract

Tetraploid complementation is often used to produce mice from embryonic stem cells (ESCs) by injection of diploid (2n) ESCs into tetraploid (4n) blastocysts (ESC-derived mice). This method has also been adapted to mouse cloning and the derivation of mice from induced pluripotent stem (iPS) cells. However, the underlying mechanism(s) of the tetraploid complementation remains largely unclear. Whether this approach can give rise to completely ES cell-derived mice is an open question, and has not yet been unambiguously proven. Here, we show that mouse tetraploid blastocysts can be classified into two groups, according to the presence or absence of an inner cell mass (ICM). We designate these as type a (presence of ICM at blastocyst stage) or type b (absence of ICM). ESC lines were readily derived from type a blastocysts, suggesting that these embryos retain a pluripotent epiblast compartment; whereas the type b blastocysts possessed very low potential to give rise to ESC lines, suggesting that they had lost the pluripotent epiblast. When the type a blastocysts were used for tetraploid complementation, some of the resulting mice were found to be 2n/4n chimeric; whereas when type b blastocysts were used as hosts, the resulting mice are all completely ES cell-derived, with the newborn pups displaying a high frequency of abdominal hernias. Our results demonstrate that completely ES cell-derived mice can be produced using ICM-deficient 4n blastocysts, and provide evidence that the exclusion of tetraploid cells from the fetus in 2n/4n chimeras can largely be attributed to the formation of ICM-deficient blastocysts.

Figure 1. Mouse tetraploid blastocysts are grouped into two types.

(A) Schematic illustration of tetraploid embryo generation. The two blastomeres of 2-cell stage diploid (2n) embryos are electrofused into one large blastomere thus doubling the DNA content to tetraploid (4n) in the embryos. The resulting 4n embryos can normally develop to blastocysts and are classified into two groups by the presence (type a) or absence (type b) of an ICM. (B) The ICM in 4n blastocysts of Oct4-EGFP embryos can be visualized by expression of EGFP in the resulting 4n blastocysts, and are classified into type a or type b under the fluorescence microscope. (C) Confocal images of diploid and tetraploid type a and tetraploid type b blastocysts, images are full projections of 20 optical sections. Embryos were stained with antibodies of CDX2 (staining the trophoblast) and OCT4 (staining the ICM). Both the diploid and tetraploid type a blastocysts showed ICM in the embryos, whereas the tetraploid type b blastocysts lacked an ICM. Arrow indicates the ICM. Scale bar: 50 µm.

Figure 2. Cell number of early embryos determines the formation of type a and type b blastocysts.

(A) Schematic illustration of how increasing or decreasing one blastomere at the 4-cell stage 4n embryos significantly affects type a and type b blastocyst formation. 4C-1: Removal of one blastomere at the 4-cell stage, the ratio of type a and type b blastocysts are 40% and 60% (χ²-test, P<0.01), respectively. 4C+1: Injection of an additional tetraploid blastomeres at the 4-cell stage, the ratio of type a and type b blastocysts are 92% and 8% (χ²-test, P<0.01), respectively. The ratio of type a and type b blastocysts in normal 4n embryos are 56% and 44%. (B) Average cell numbers of expanded blastocysts. 4n(4C):The average cell number in type b blastocysts from normal 4n embryos is lower than in type a blastocysts (t-test, P<0.01). 4n(4C-1): The cell number in the blastocysts from embryos with one blastomere removed at the 4-cell stage (type a and type b were not grouped). 4n(4C+1): The cell number in the blastocysts from embryos where one 4n blastomere was added at the 4-cells stage. 2n: Normal expanded 2n blastocysts. (C) A type b 4n blastocyst from 4C-1 embryos showing the decreased cell number and missing the ICM. Scale bar: 20 µm.

Figure 3. Embryonic stem cell (ESC) lines derived from type a 4n blastocysts.

(A) Inner cell mass (ICM) outgrowth on MEFs from a type a Oct4-EGFP 4n blastocyst at the 4^th day of culture in ESDM. (B) An embryonic remnant from a type b 4n blastocyst at the 4^th day of culture in ESDM. (C, D) ESCs derived from both 2n embryos and type a 4n embryos. Note the larger cell size for 4nESCs (D) than 2nESCs (C). (E) Karyotype of 4nESCs (4n = 80). (F) Efficiency of ESC line derivation for 4n blastocysts from Oct4-EGFP mouse strain. The efficiency of ESC derivation for type a 4n blastocysts is similar to that for 2n blastocysts, while the efficiency for type b 4n blastocysts is significantly decreased (P<0.01). The type a and type b embryos were identified by visualization of the Oct4-EGFP under the fluorescence microscope.

Figure 4. 4nESCs are pluripotent and contribute substantially to chimeras.

(A, B) 4nESCs express pluripotent markers, such as AP staining positive (A) and OCT4 expression (B). (C) A chimera (4n/2n) obtained by injection of 4nESCs (black coat color) into a 2n host blastocyst (white coat color). The chimera showed over 50% of 4nESCs contribution judging by the coat color. (D) A new born pup showed the contribution of 4nESCs (evidenced by the CMV promoter driven EGFP) in the 4n/2n chimera. (E) Fibroblasts from a newborn 4nESC/ICR chimera cultured in vitro, the derivatives of 4nESCs are EGFP⁺. (F) Karyotype of EGFP⁺ fibroblasts from chimera (4n = 82, (1, +1; 14, +1)). (G) Two populations of fibroblasts (EGFP⁻ and EGFP⁺). (H) The DNA content of the total population of fibroblasts stained with PI shows three peaks: 2n, 4n and 8n. (I) The DNA content of sorted EGFP⁻ fibroblasts shows two major peaks: 2n and 4n. (J) The DNA content of sorted GFP⁺ fibroblasts show two peaks: 4n and 8n. Scale bar: 200 µm.

Figure 5. ESC mice produced by tetraploid complementation using both type a and type b 4n blastocysts.

(A) Newborn pups obtained using type a and type b 4n blastocysts by tetraploid complementation. Pups from type b 4n blastocysts are frequently displaying abdominal hernia (arrow indicated), while pups from type a 4n blastocyst are all normal. (B) A litter of ESC pups (white coat color) produced using type a 4n blastocysts (black coat color); one pup (arrow indicated) displayed substantial contribution of cells from the host 4n embryo (over 20% of contribution from the host embryo judging by the coat color). (C) The efficiency to obtain ESC mice from type a or type b 4n blastocysts is not significantly different (χ²-test, P = 0.13). (D)The frequency of herniated pups using type b blastocysts is significantly higher than using type a blastocysts (χ²-test, P<0.01). (E) A model for tetraploid complementation illustrates ESC mice from type a 4n blastocysts are possibly 2n/4n chimeras, whereas ESC mice from type b 4n blastocysts could be pure ESC-derived.