Letter to the Editor Mice cloned from white adipose tissue-derived cells

Dear Editor, Somatic cells can be reprogrammed into totipotent or pluripotent state by nuclear transfer (NT) or the ‘Yamanaka method’. White adipose tissue (WAT) is a source of readily accessible donor cells that provide virtually unlimited cells for reprogramming. Recently, freshly isolated white adipose tissue-derived cells (ADCs) from humans or mice have been successfully reprogrammed into induced pluripotent stem cells (iPSCs) by ‘Yamanaka factors’ (Sun et al., 2009; Sugii et al., 2010). In contrast, cultured mouse ADCs were not feasible donors for NT, most likely due to chromosome instability that might occur during in vitro culture (Qin et al., 2009). Therefore, it remains unknown whether genomic stability sustains in freshly isolated mouse ADCs that can be used as donors for NT. Here, we report that cloned mice can be produced from ADCs and cloning efficiency can be dramatically improved by replacing the trophoblast cells in the cloned blastocysts with cells from tetraploid embryos. Moreover, the cloning efficiencies are consistently higher for Lin cells than for CD45+ cells. Our results indicate that ADCs are an easily obtainable cell source that can be used as donors for studying somatic reprogramming induced by oocytes. WAT is composed of heterogeneous cell populations, including multipotent progenitor cells and differentiated cells. According to cell lineage markers, ADCs can be separated into two groups: a lineage-positive (Lin+) population for endothelial cells (CD31), erythrocytes (Ter119), and haematopoietic cells (CD45), and a lineage-negative (Lin) population for the remaining cells primarily composed of progenitor cells including preadipocytes, prevascular cells, and mesenchymal stem cells (MSCs) (Rodeheffer et al., 2008). To test the reprogramming ability of Lin and Lin+ cells, adult male B6D2F1 mice were used for the isolation of different ADCs by fluorescence-activated cell sorting (FACS) as described previously (Figure 1A) (Rodeheffer et al., 2008). Briefly, the stromal vascular fraction (SVF) was isolated by digestion and centrifugation, and was incubated with conjugated antibodies against the cell-surface markers expressed by Lin+ cells, including CD31, CD45, and Ter119 (Supplementary Figure S1A, P1 zone). First, CD31 and Ter119 doublenegative cells (Supplementary Figure S1A, P2 zone) or CD45 and Ter119 doublenegative cells (Supplementary Figure S1A, P5 zone) were enriched in the SVF by removing CD31+ and Ter119+ cells or CD45+ and Ter119+ cells, respectively. Lin cells (Supplementary Figure S1A, P3 or P6 zone) were further enriched by removing CD45+ cells from CD31 and Ter119 doublenegative cells (Supplementary Figure S1A, P2 zone) or by removing CD31+ cells from CD45 and Ter119 double-negative cells (Supplementary Figure S1A, P5 zone). Haematopoietic (CD45+) (Supplementary Figure S1A, P4 zone) and endothelial (CD31+) (Supplementary Figure S1A, P7 zone) cells were collected for further analyses. To test the purity of FACS-enriched Lin, CD31+, and CD45+ cells, RT–PCR (Supplementary Table S1) was performed to analyse the expression of marker genes in these cells. As expected, surface markers CD31 and CD45 expressed only in CD31+ and CD45+ cells, respectively. By contrast, Lin cells expressed neither CD31 nor CD45 (Supplementary Figure S1B). Furthermore, flow analysis showed that Lin cells highly expressed CD140A, CD140B, CD105, CD13, and Sca-1, putative markers for prevascular cells and MSCs, and CD34, a marker for preadipocytes (Supplementary Figure S1C). To test the multipotency of Lin cells, we performed in vitro differentiation assays according to a previous report (Qin et al., 2009) and found that Lin cells could differentiate into chondrogenic, osteogenic, and adipogenic cells (Supplementary Figure S1D). Taken together, these data support the hypothesis that different cell populations, including multipotent cells (Lin cells) and differentiated cells (CD45+ and CD31+ cells), can be isolated from adipose tissue and sorted using cell-surface markers. To compare the reprogramming efficiency of multipotent and differentiated cells from adipose tissue by NT, freshly isolated Lin cells and CD45+ cells were respectively used for NT. We chose CD45+ cells to represent the differentiated cell type because the number of CD45+ cells in SVF was much higher than that of CD31+ cells (Supplementary Figure S1A). Mature oocytes were collected from superovulated wild-type mice (B6D2F1). After enucleation of oocytes, donor nuclei from Lin or CD45+ cells were transferred by piezo-actuated microinjection. The reconstructed oocytes were then cultured in vitro in standard embryo culture medium (KSOM) until the blastocyst stage. The rate of in vitro development of reconstructed oocytes into blastocysts was significantly higher for Lin cells (termed Lin embryos) (Figure 1B) than for CD45+ cells (CD45+ embryos) (55.3% vs. 23.3%; P , 0.01) (Supplementary Table S1). Similarly, in vivo development was higher for Lin cells than for CD45+ cells (2.3% vs. 0% per transferred blastocyst). The cellcycle stage is a critical factor that affects cloning efficiency (Wilmut et al., 1997). Flow analysis indicated that both freshly isolated Lin cells and CD45+ cells were at the G0/G1 stage, which has been demonstrated to be the optimal cell-cycle stage for NT. A total of four liveborn cloned mice were produced from freshly isolated Lin cells in this study (Figure 1C and Supplementary Table S1). The cloned pups had a normal weight at birth, but their placentas weighed 2to 3-fold more than normal placentas, 348 | Journal of Molecular Cell Biology (2013), 5, 348–350 doi:10.1093/jmcb/mjt019 Published online June 10, 2013