Differentiated cells are more efficient than adult stem cells for cloning by somatic cell nuclear transfer Li-Ying Sung 1,6 , Shaorong Gao 1,6,7 , Hongmei Shen 2,7 , Hui Yu 2 , Yifang Song 2 , Sadie L Smith 1 , Ching-Chien Chang 1 , Kimiko Inoue 1,7 , Lynn Kuo 3 , Jin Lian 4 , Ao Li 5 , X Cindy Tian 1 , David P Tuck 5 , Sherman M Weissman 4 , Xiangzhong Yang 1 & Tao Cheng 2 Since the creation of Dolly via somatic cell nuclear transfer (SCNT) 1 , more than a dozen species of mammals have been cloned using this technology 2 . One hypothesis for the limited success of cloning via SCNT (1%–5%) 3 is that the clones are likely to be derived from adult stem cells 4 . Support for this hypothesis comes from the findings that the reproductive cloning efficiency for embryonic stem cells is five to ten times higher than that for somatic cells as donors 5,6 and that cloned pups cannot be produced directly from cloned embryos derived from differentiated B and T cells or neuronal cells 7–10 . The question remains as to whether SCNT-derived animal clones can be derived from truly differentiated somatic cells. We tested this hypothesis with mouse hematopoietic cells at different differentiation stages: hematopoietic stem cells, progenitor cells and granulocytes. We found that cloning efficiency increases over the differentiation hierarchy, and terminally differentiated postmitotic granulocytes yield cloned pups with the greatest cloning efficiency. The mouse hematopoietic stem cell (HSC) is phenotypically the best- defined tissue stem cell among all the adult stem cell types 11 . Using our standard methods, we isolated mouse hematopoietic cells at different differentiation stages: namely, HSCs, hematopoietic progeni- tor cells (HPCs) and granulocytes 12,13 (Fig. 1a). We chose the BDF1 (C57BL/6J  DBA/2) hybrid mouse strain 14 for this study primarily because it has proven successful in mouse cloning 10,15 . Mouse marrow HSCs are contained in the cell population negative for blood cell lineage markers (Lin – ) and positive for c-Kit and Sca-1 (LKS). LKS cells are further divided into CD34 – and CD34 + subsets, which represent the long-term repopulating HSC (LT-HSC) and short- term repopulating HSC (ST-HSC) populations, respectively 16 . Between HSCs and differentiated cell populations, there are pools of committed, intermediate HPCs in Lin – c-Kit + populations that are largely negative for Sca-1 (LKS – ). Our isolation and purification efficiencies for all these hematopoietic cell subsets by flow cytometry are 97%–98% (Fig. 1b). Mature granulocytes are terminally differ- entiated postmitotic cells with distinct morphological features of segmented nuclei and cytoplasmic granules. We isolated granulocytes prospectively by flow cytometry for Gr-1 high expression in the gran- ulocytic scatter region, and our isolation achieved 99.4% purity, as assessed by morphological criteria (Fig. 1c). We compared the cloning efficiency for preimplantation develop- ment in vitro using highly purified HSCs, HPCs and granulocytes as donor cells for SCNT. We froze the isolated hematopoietic cell populations (Fig. 1) in liquid nitrogen for shipment and later used them for SCNT without in vitro culture. Cell viability after thawing was 90% by trypan blue staining (data not shown). Only viable cells with healthy, smooth and intact membranes were selected as nuclear donor cells for SCNT. Preimplantation development of cloned embryos reconstructed with different hematopoietic cells is shown in Table 1. We were surprised to find that the SCNT cloning efficiency of morula and blastocyst development rate was lowest for LT-HSCs (that is, 4%) and that the cloning efficiency increased with the stage of differentiation to 8% for ST-HSCs, 11% for HPCs and 35% for granulocytes (Table 1). We performed a total of 1,828 nuclear transfers from HSCs and HPCs, and most nuclear transfer–cloned embryos from those cells arrested at the two- to four-cell stage. Consistent with a previous report 17 , the HSC clones showed very limited developmental potential in vitro, thereby precluding us from subsequently performing sufficient embryo transfers with morulae or blastocysts from HSC donors (data not shown). Given that the highest rate of blastocyst development among the hematopoietic cell subsets tested was in granulocytes, we further explored the possibility of producing cloned pups from granulocyte- derived cloned embryos. In total, we performed 1,368 SCNTs from granulocytes in two experiments; 34.5%–38.9% of the cloned embryos Received 19 May; accepted 5 September; published online 1 October 2006; doi:10.1038/ng1895 1 Center for Regenerative Biology and Department of Animal Science, University of Connecticut, Storrs, Connecticut 06269, USA. 2 Cancer Stem Cell Program, University of Pittsburgh Cancer Institute and Department of Radiation Oncology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213, USA. 3 Department of Statistics, University of Connecticut, Storrs, Connecticut 06269, USA. 4 Department of Genetics and 5 Department of Pathology, Yale University School of Medicine, New Haven, Connecticut 06510, USA. 6 These authors contributed equally to this work. 7 Present addresses: National Institute of Biological Sciences, Beijing, P.R. China (S.G.); The Thomas E. Starzl Transplantation Institute, University of Pittsburgh, Pittsburgh, Pennsylvania, USA (H.S.) and RIKEN Bioresource Center, Tsukuba, Ibaraki 305-0074, Japan (K.I.). Correspondence should be addressed to X.Y. (xiangzhong.yang@uconn.edu) or T.C. (chengt@upmc.edu). NATURE GENETICS VOLUME 38 [ NUMBER 11 [ NOVEMBER 2006 1323 LETTERS © 2006 Nature Publishing Group http://www.nature.com/naturegenetics