Stem Cells on Patrol

be interesting to determine whether the transition to a mesenchymal fate provides an advantage to cancer cells enabling them to escape entosis, or whether entosis is deregulated in cancers. As with the initial description of a new phenomenon, which often raises more questions than it provides answers, defining the process of entosis and the genes that control it will be needed to establish its relevance in vivo.

References Abodief, W.T., Dey, P., and Al-Hattab, O. (2006). Cytopathology 17, 304–305. Debnath, J., Mills, K.R., Collins, N.L., Reginato, M.J., Muthuswamy, S.K., and Brugge, J.S. (2002). Cell 111, 29–40. Gilmore, A.P. (2005). Cell Death Differ. 12 (Suppl 2), 1473–1477. Karantza-Wadsworth, V., Patel, S., Kravchuk, O., Chen, G., Mathew, R., Jin, S., and White, E. (2007). Genes Dev. 21, 1621–1635.

Lindsten, T., Ross, A.J., King, A., Zong, W.X., Rathmell, J.C., Shiels, H.A., Ulrich, E., Waymire, K.G., Mahar, P., Frauwirth, K., et al. (2000). Mol. Cell 6, 1389–1399. Mailleux, A.A., Overholtzer, M., Schmelzle, T., Bouillet, P., Strasser, A., and Brugge, J.S. (2007). Dev. Cell 12, 221–234. Nelson, C.M., and Bissell, M.J. (2006). Annu. Rev. Cell Dev. Biol. 22, 287–309. Overholtzer, M., Mailleux, A.A., Mouneimne, G., Normand, G., Schnitt, S.J., King, R.W., Cibas, E.S., and Brugge, J.S. (2007). Cell, this issue.

Stem Cells on Patrol Robert S. Welner1 and Paul W. Kincade1,*

Immunobiology and Cancer Program, Oklahoma Medical Research Foundation, 825 N.E. 13th Street, Oklahoma City, OK 73104, USA *Correspondence: [email protected] DOI 10.1016/j.cell.2007.11.010

1

Hematopoietic stem cells (HSCs) exist in the bone marrow and circulate in the blood. In this issue, Massberg et al. (2007) report that HSCs also travel through the lymphatic system. Furthermore, migration of HSCs—which express Toll-like receptors—allows the recognition of pathogenic molecules in peripheral tissues thereby promoting the local generation of innate immune cells at the site of infection. Hematopoietic stem cells (HSCs) were once believed to avoid the ravages of age by residing in bone marrow, where they generate cells of nearly all hematopoietic lineages. Yet, HSCs are also present in blood and other tissues. Studies that measure the numbers of HSCs in blood reveal that nearly all HSCs in the bone marrow periodically enter the cell cycle (Cheshier et al., 1999; Wright et al., 2001). In this issue, Massberg, von Andrian, and their colleagues chart new patrol routes for HSCs, deduce how their migration is controlled, and establish a raison d’être outside of the bone marrow (Massberg et al., 2007). There is recent support for the long-held view that HSCs reside in association with osteoblasts near trabecular bone—the tissue in the inner cavities of bones (Adams

and Scadden, 2006). This niche is thought to represent a nurturing environment, providing signals for HSC survival, relative quiescence, and self renewal (Figure 1). However, methods for identifying HSCs have progressively improved, and we now know that larger numbers of HSCs can be found in the more centrally located perivascular regions of bones (Kiel et al., 2005). Although the bone marrow is still considered to be the principal site for blood cell formation, spleen and liver are also important sites for this process during fetal development. Moreover, blood cell formation in these tissues can be transiently reactivated (termed extramedullary hematopoiesis) under conditions of unusual demand for blood cells such as during a parasitic infection or during myelofibrosis.

842  Cell 131, November 30, 2007 ©2007 Elsevier Inc.

Although mobilization of HSCs for therapeutic transplantation is being extensively studied (Winkler and Levesque, 2006), mechanisms associated with normal egress into the circulation are poorly understood. It is likely that cell adhesion molecules, proadhesive chemokines, crowding, neutrophilderived proteases, and innervation of niches by the sympathetic nervous system all determine whether HSCs remain sessile or are induced to migrate. The blood of a mouse contains between 100 and 400 HSCs at any one time, but they reside there for only seconds, and some eventually return to the bone marrow (Wright et al., 2001). Thus, there may be continual low-level exchange of HSCs between hematopoietic niches in different bones. Many studies suggest that export of stem and progenitor cells from bone marrow can increase dramatically during infection.

Figure 1. Migration of Hematopoietic Stem and Progenitor Cells (1) Many stem cells reside in association with osteoblasts in trabecular bone. (2) An even greater number are close to centrally located vascular sinuses, but little is known about exchange between these two locations. (3) Pathogen products and cytokines released during infection can mobilize hematopoietic stem and progenitor cells (HSPCs), which migrate out of the marrow into blood, but small numbers exit under normal circumstances. HSPCs remain in the circulation for only seconds but reside in tissues such as the lungs, liver, and spleen for at least 36 hr. (4) Their stay is extended by recognition of microbial or viral products by the Toll-like receptors (TLR) that they express. Dendritic and myeloid cells are produced locally from HSPCs in response to particular TLR ligands, and these effectors of the innate immune system presumably fight infections and promote tissue repair. (5) Unstimulated HSPCs use their sphingosine-1-phosphate receptors (such as S1P1) to recognize steep gradients of sphingosine-1-phosphate and to enter the lymph (blue). (6) HSPCs quickly transit through one or more lymph nodes before returning to the blood via the thoracic duct. (7) At least some stem cells return to marrow niches, but it is not known if they are affected by the journey.

Intravenously administered HSCs quickly go to the spleen and some reenter the blood for transit to a variety of tissues (Wright et al., 2001). In the new study, von Andrian and colleagues (Massberg et al., 2007) now show that lymph and lymph nodes represent another major migration route for HSCs (Figure 1). They demonstrate that HSCs reside for a time in nonhematopoietic tissues before transiting through the lymph nodes, eventually re-entering the blood circulation via the thoracic duct. Gradients of sphingosine-1-phosphate regulate egress of lymphocytes from the thymus, spleen, and lymph nodes (Pappu et al., 2007). Massberg et al. (2007) show that recognition of sphingosine-1-phosphate by receptors on the HSCs controls their movement from tissues into the lymph. Their work also suggests that the route of HSC migration is different from that of lymphocytes. HSCs enter lymph via afferent lymphatics rather than high endothelial venules and spend little time in lymph

nodes before re-entering blood via the thoracic duct. Indeed, they even recirculate via lymph in mice that lack lymph nodes. All of these findings suggest that there is migration of HSCs, but what is the purpose? HSC and lineagespecified progenitors express functional Toll-like receptors (TLRs) (Nagai et al., 2006). In humans as well as flies, TLRs act as sentinels in cells that express them by recognizing foreign molecules such as those expressed by pathogens. TLR ligands such as the bacterial outer membrane component lipopolysaccharide promote entry of quiescent HSCs into the cell cycle, lower the cytokine requirements for differentiation of myeloid progenitors, and direct lymphoid progenitors to take on a dendritic cell fate (Nagai et al., 2006; R.W., R. Pelayo, and P.K., unpublished data). Lipopolysaccharide injected intravenously quickly finds its way to HSCs in the bone marrow. This suggests that stem/progenitor cells

could be stimulated at that site by circulating pathogen products, resulting in generation of more cells of the innate immune system. Massberg et al. (2007) now show that HSCs can also address the threat of infection elsewhere in the body. Coinjection of lipopolysaccharide and HSCs beneath the kidney capsule simulated an infection site; this TLR ligand encouraged retention of hematopoietic cells in peripheral tissues and stimulated their proliferation and the preferential generation of dendritic cells. These results indicate that migration enables HSCs to recognize pathogens at infection sites and rapidly produce innate immune effector cells. These exciting observations blur the distinction between central versus peripheral hematopoiesis and raise many interesting questions. For example, does self renewal of HSCs also occur outside the bone marrow, or do HSCs expend their expansion potential while attempting to counter threats? Massberg et al. (2007) demonstrate in vivo what has been previously observed in culture (Nagai et al., 2006), that lipopolysaccharide stimulates dendritic cell production. There are many functionally specialized types of dendritic cells, and recent findings suggest that they may have different developmental origins (Wu and Liu, 2007). Patterns of differentiation depend on location of progenitors and exposure to growth and differentiation factors. New observations suggest that TLR ligands represent additional environmental cues for inducing dendritic cell formation, and the nature of the dendritic cells produced depends on which pathogen product is involved (Nagai et al., 2006; R.W., R. Pelayo, and P.K., unpublished data). Fate-mapping models need to be developed that can be used to trace dendritic cell genealogies under normal and disease circumstances. Given the scarcity of cells involved in this study, many of the experiments and inferences pertain to myeloid progenitors or to a cell fraction comprising HSCs and progenitors together rather than to a pure HSC population. Even highly purified HSCs are heterogeneous, and subsets of HSCs appear to be intrinsically biased toward particu-

Cell 131, November 30, 2007 ©2007 Elsevier Inc.  843

lar blood cell lineages (Muller-Sieburg and Sieburg, 2006). Could this reflect or determine their migration routes? Pathogen products can recruit additional HSCs to tissues, but there must be antagonistic mechanisms and ways to stimulate their self renewal. Otherwise, chronic infections would exhaust our stem cell reserve. HSCs acting as scouts in peripheral tissues would not be expected to be constrained by molecules that inhibit their proliferation in the bone marrow. Moreover, there are age-related increases in HSC numbers and skewing of myeloid-to-lymphoid differentiation (Rossi et al., 2007). It will be interesting to determine whether migrating HSCs are particularly susceptible to aging. The seminal work of

Massberg, von Andrian, and their colleagues is certain to catalyze further exploration of how migration of stem cells protects and replaces tissues.

Muller-Sieburg, C.E., and Sieburg, H.B. (2006). Cell Cycle 5, 394–398.

References

Pappu, R., Schwab, S.R., Cornelissen, I., Pereira, J.P., Regard, J.B., Xu, Y., Camerer, E., Zheng, Y.W., Huang, Y., Cyster, J.G., and Coughlin, S.R. (2007). Science 316, 295–298.

Adams, G.B., and Scadden, D.T. (2006). Nat. Immunol. 7, 333–337. Cheshier, S.H., Morrison, S.J., Liao, X., and Weissman, I.L. (1999). Proc. Natl. Acad. Sci. USA 96, 3120–3125. Kiel, M.J., Yilmaz, O.H., Iwashita, T., Yilmaz, O.H., Terhorst, C., and Morrison, S.J. (2005). Cell 121, 1109–1121. Massberg, S., Schaerli, P., Knezevic-Maramica, I., Kollnberger, M., Tubo, N., Moseman, E.A., Huff, I.V., Junt, T., Wagers, A.J., Mazo, I.B., and Von Andrian, U.H. (2007). Cell, this issue.

Nagai, Y., Garrett, K.P., Ohta, S., Bahrun, U., Kouro, T., Akira, S., Takatsu, K., and Kincade, P.W. (2006). Immunity 24, 801–812.

Rossi, D.J., Bryder, D., and Weissman, I.L. (2007). Exp. Gerontol. 42, 385–390. Winkler, I.G., and Levesque, J.P. (2006). Exp. Hematol. 34, 996–1009. Wright, D.E., Wagers, A.J., Gulati, A.P., Johnson, F.L., and Weissman, I.L. (2001). Science 294, 1933–1936. Wu, L., and Liu, Y.J. (2007). Immunity 26, 741–750.

cis-Regulatory Elements within the Odorant Receptor Coding Region Lillian C. Merriam1 and Andrew Chess1,*

Center for Human Genetic Research and Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Simches Research Building, 185 Cambridge Street, Boston, MA 02114, USA *Correspondence: [email protected] DOI 10.1016/j.cell.2007.11.016

1

Complex regulatory mechanisms lead to the expression in each olfactory neuron of one allele of only one of the 1000 odorant receptor (OR) genes. In this issue, Nguyen et al. (2007) provide evidence that regulatory elements residing within the coding region of OR genes are involved in the singularity of OR gene expression. Although there are over 1000 odorant receptor (OR) genes in the mouse genome, in any given olfactory sensory neuron, a functional protein is stably expressed from only one allele of one OR gene (Buck and Axel, 1991; Chess et al., 1994; Malnic et al., 1999). What is the mechanism that allows a single allele of a single odorant receptor gene to be expressed in each olfactory neuron? The large size of the OR gene family and the distribution of family members across most chromosomes make it difficult to fathom how the precision of this exclusivity is achieved. Findings by

Nguyen et al. (2007) now suggest that elements within the coding region itself mediate exclusivity of OR gene expression. Previous work has reported the exciting possibility that there is a single expression site for OR genes in the nucleus, which would help to explain the singularity of OR gene expression (Lomvardas et al., 2006). Chromosome capture and fluorescence in situ hybridization (FISH) experiments indicated that a conserved 2 kb region near an odorant receptor cluster on mouse chromosome 14, called the H region (Lane et al., 2002;

844  Cell 131, November 30, 2007 ©2007 Elsevier Inc.

Serizawa et al., 2003), was physically associated with the OR gene chosen for expression irrespective of its genomic location (Lomvardas et al., 2006). This led to the idea that the H region was a key regulator of the entire OR gene family. Doubt was cast on this conclusion, however, by experiments knocking out the H region in mice and demonstrating that numerous scattered OR genes maintain seemingly normal expression (Fuss et al., 2007). In this knockout mouse, it was only the nearest OR genes that had their expression extinguished. Yet,