Abstract
We recently demonstrated human T cell, B cell, dendritic cell, and natural interferon producing cell development and consecutive formation of primary and secondary lymphoid organs in Rag2−/−gc−/− mice, transplanted as newborns intra-hepatically (i.h.) with human CD34+ cord blood cells (Traggiai et al., Science 2004). Although these mice support high levels of human cell engraftment and continuous T and B cell formation as well as CD34+ cell maintenance in bone marrow over at least six month, the frequency of secondary recipient reconstituting human hematopoietic stem and progenitor cells within the CD34+ pool declines over time. Also, although some human immune responses are detectable upon vaccination with tetanus toxoid, or infection with human lymphotropic viruses such as EBV and HIV, these responses are somewhat weak compared to primary human responses, and are inconsistent in frequency. Thus, some factors sustaining human hematopoietic stem cells in bone marrow and immune responses in lymphoid tissues are either missing in the mouse environment, or are not cross-reactive on human cells.
Human mesenchymal stem cells (MSCs) replicate as undifferentiated cells and are capable to differentiate to multiple mesenchymal tissues such as bone, cartilage, fat, muscle, tendon, as well as marrow and lymphoid organ stroma cells, at least in vitro (e.g. Pittenger et al., Science 1999). Moreover, it was shown that MSCs maintain CD34+ cells to some extend in vitro, and engraft at low frequency upon transplantation into adult immunodeficient mice or fetal sheep as detected by gene transcripts. We thus postulated that co-transplantation of cord blood CD34+ cells and MSCs into newborn mice might lead to engraftment of both cell types, and to provision of factors supporting CD34+ maintenance and immune system function. MSCs were isolated and expanded by plastic adherence in IMDM, supplemented with FCS and cortisone (first 3 weeks) from adult bone marrow, cord blood, and umbilical vein. To test their potential to support hemato-lymphopoiesis, MSCs were analyzed for human hemato-lymphotropic cytokine transcription and production by RT-PCR and ELISA, respectively. MSCs from all sources expressed gene-transcripts for IL-6, IL-7, IL-11, IL-15, SCF, TPO, FLT3L, M-CSF, GM-CSF, LIF, and SDF-1. Consistently, respective cytokines were detected in supernatants at the following, declining levels (pg/ml): IL-6 (10000-10E6) > SDF-1 > IL-11 > M-CSF > IL-7 > LIF > SCF > GM-CSF (0–450), while FLT3L and TPO were not detectable by ELISA. Upon i.h. transplantation of same passage MSCs (1X10E6) into sublethally irradiated (2x2 Gy) newborn Rag2−/−gc−/− mice, 2-week engraftment was demonstrated by species specific b2m-RT-PCR in thymus, spleen, lung, liver and heart in n=7 and additionally in thymus in n=3 out of 13 animals analyzed. Equally, GFP-RNA transcripts were detectable in the thymus for up to 6 weeks, the longest time followed, upon co-transplantation of same source CD34+ cells and retrovirally GFP-transduced MSCs in n=2 out of 4 animals. Further engraftment analysis of ongoing experiments will be presented.
Overall, these results demonstrate that human MSC produce hemato-lymphoid cytokines and engraft in newborn transplanted Rag2−/−gc−/− mice, at least at early time-points analyzed. This model thus might allow studying hematopoietic cell and MSC-derived cell interaction, and might serve as a testing system for MSC delivered gene therapy in vivo.
In vitro analysis of integrin expression in stem cells from bone marrow and cord blood during chondrogenic differentiation