Key words: T cell development, yolk sac, hematopoietic stem cells, thymus. Introduction. Studies on the origin and the nature of the stem cells of the T-cell ...
Development 113, 1315-1323 (1991) Printed in Great Britain © The Company of Biologists Limited 1991
1315
In vitro development of murine T cells from prethymic and preliver embryonic yolk sac hematopoietic stem cells
CHIH-PIN LIU and ROBERT AUERBACH Center for Developmental Biology, University of Wisconsin, Madison, WI 53706, USA
Summary
Mature T cells are derived from prethymic stem cells, which arise at one or more extrathymic sites and enter and differentiate in the thymus. The nature of these prethymic stem cells is a critical factor for the formation of the T-cell repertoire. Although the bone marrow of adult mice can provide such stem cells, their origin during murine embryogenesis is still undetermined. Among potential sites for these progenitor cells are the fetal liver and the embryonic yolk sac. Our studies focus on the yolk sac, both because the yolk sac appears earlier than any other proposed site, and because the mammalian yolk sac is the first site of hematopoiesis. Although it has been shown that the yolk sac in midgestation contains stem cells that can enter the thymic rudiment and differentiate toward T-cell lineage, our aim was to analyze the developmental potential of cells in the yolk sac from earlier stages, prior to the formation of the liver and any other internal organ. We show here that the yolk sac from 8- and 9-day embryos (2-9 and 13-19 somites, respectively) can reconstitute alymphoid congenic fetal thymuses and acquire mature
T-cell-specific characteristics. Specifically, thymocytes derived from the early embryonic yolk sac can progress to the expression of mature T lymphocyte markers including CD3/T-cell receptor (TCR), CD4 and CD8. In contrast, we have been unable to document the presence of stem cells within the embryo itself at these early stages. These results support the hypothesis that the stem cells capable of populating the thymic rudiment originate in the yolk sac, and that their presence as early as at the 2- to 9-somite stage may indicate that prethymic stem cells found elsewhere in the embryo at later times may have been derived by migration from this extraembryonic site. Our experimental design does not exclude the possibility of multiple origins of prethymic stem cells of which the yolk sac may provide the first wave of stem cells in addition to other later waves of cells.
Introduction
has been proposed, including the embryonic yolk sac, fetal liver and omentum (Auerbach et al. 1981; Owen and Jenkinson, 1981; Moore and Owen, 1967; Metcalf and Moore, 1971; Eren et al. 1987a, b; Kubai and Auerbach, 1983; Toles et al. 1989; Liu and Auerbach, 1991). Previous studies have shown that the yolk sac from as early as 10-day-old embryos contains stem cells capable of entering the thymus, and that they can proliferate in the thymic microenvironment and differentiate there toward the T-cell lineage (Eren et al. 1987a, b; Liu and Auerbach, 1991). However, at day 10, the liver already contains hematopoietic stem cells and the omental rudiment has begun to form from the gastroepiploic mesentery (Rugh, 1968). We wished to determine whether we might be able to locate stem cells before there is even rudimentary development of the liver or omentum, i.e. whether prethymic hematopoietic stem cells of T cells might be present in the yolk sac or embryonal body at earlier times in gestation. We report here that syngeneic as well as allogeneic day 8 and day 9
Studies on the origin and the nature of the stem cells of the T-cell lineage are important to our understanding of the establishment of the T-cell repertoire. It is known that mature T cells are derived from prethymic hematopoietic stem cells which arise at one or more extrathymic sites and then enter, proliferate and differentiate in the thymus, acquiring T-cell-specific characteristics, such as the expression of CD3/TCR molecules (Auerbach et al. 1981; Owen and Jenkinson, 1981; Moore and Owen, 1967; Metcalf and Moore, 1971; Ikuta et al. 1990; Spangrude et al. 1988). It has been shown that adult and embryonic hematopoietic stem cells may differ in their potential for giving rise to various T-cell subsets, such as TCR Vy3-bearing cells, in the thymus (Ikuta et al. 1990). While the bone marrow of adult mice is clearly a major site for such stem cells (Spangrude et al. 1988), it is still not clear where the stem cells arise during mammalian embryogenesis. Moreover, more than one such embryonic site
Key words: T cell development, yolk sac, hematopoietic stem cells, thymus.
1316
C.-P. Liu and R. Auerbach
yolk sac cells can reconstitute alymphoid Thyl congenic thymus lobes in vitro and that these yolk-sac-derived cells acquire T-cell-specific characteristics critical to T-cell-mediated immune functions, including the cell surface expression of CD3/TCR, CD4 and CD8 molecules, in a manner similar to what is seen in normal in situ fetal thymic development. In contrast to our findings using the cells from extraembryonic yolk sac, we were not able to effect reconstitution of thymic stem-cell-depleted organ cultures using cells from the embryo itself at these early stages of mammalian development. The results of these studies suggest that the prethymic stem cells capable of seeding the thymic rudiment may arise in the yolk sac and then migrate to elsewhere in the embryo. Materials and methods Mice BALB/cAu (H-2d, Thyl.2), C57BL/6Au (H-2b, Thyl.2) and C57BL/6KaAu (H-2b, Thyl.l) were maintained in our own facilities. Embryos at various stages of gestation were obtained from mice using timed matings, with the appearance of a vaginal plug counted as day 0. Thymus lobes used for reconstitution were obtained from day 14 C57BL/6KaAu embryos. The embryos, obtained from BALB/cAu or C57BL/6Au, were washed extensively to eliminate the possibility of contamination by maternal blood. Yolk sacs and embryonal bodies were carefully dissected from day 8 or 9 embryos and were mechanically dissociated to prepare single cell suspensions. The gestation age of the embryo used for experiments was further determined on the basis of the number of somites. Day 8 embryos with 2 to 9 somites and day 9 embryos with 13 to 19 somites were used in reconstitution studies. Yolk sacs and embryonal bodies were isolated either from allogeneic (BALB/c) or from syngeneic (C57BL/6) strains of mice. The viability of the isolated single cell suspensions was always greater than 80 % right before their use for thymus reconstitution as determined by trypan blue dye exclusion. Antibodies Antibodies used in the studies included: anti-Thyl.l [from American Type Culture Collection (ATCC)]; anti-Thyl.2 either labeled with FITC or biotinylated, anti-CD4 (L3T4) labeled with phycoerythrin (PE), anti-CD8 (Lyt2) either labeled with FITC or biotinylated (all from Becton-Dickinson); anti-CD3 (145-2C11; Leo et al. 1987), anti-TCR o//3 (H57-597; Kubo et al. 1989) and anti-TCR y/
11 .
ft
1
IV
IV
CD3
CD3
D.
>
•
1 1 • IV
CD3
CD3
E.
\
- ,:r.f7Cn« : : ' CONTROL
CONTROL
Fig. 4. Two-color flow cytometric analysis of cells obtained from allogeneic (BALB/c) day 8 yolk sac (A and C) and embryonal body (B and D) cells reconstituted thymuses cultured for 15 days. In A and B, cells were stained with Thy 1.2 vs. CD3. C and D are the controls for A and B, respectively. Quadrants shown were selected to achieve optimal resolution between autofluorescence and specific staining (A vs. C; B vs. D).
Yolk sac origin of murine T cells 1321
B.
C\J :.
THY1.1
C.
•-
•.
CD3
D.
C\J
c\j
X
1
•t 111
.
THY1.1
be segregated and present in both the yolk sac and the embryonal body, the cells in each compartment may have differential capability to differentiate toward the T-cell lineage in the in vitro thymus reconstitution system. Thymus reconstitution studies require that the donor stem cells are capable of entering the thymus and can go on to differentiate there. Thus, where we failed to obtain stem cell repopulation (e.g. with intraembryonic cells), it can be argued that the stem cells in the embryonal body are not able to enter the thymus, that they do enter and differentiate but are at a competitive disadvantage and displaced by the time of assay or delayed in their development, or alternatively that the alymphoid thymus microenvironment is not supportive of their proliferation and differentiation even if they can enter the thymus. Studies using direct injection of the donor cells into the thymus may help to establish or negate these possibilities. Third, it may be that not all T cells (T-cell subsets; different TCR variable gene usage) arise at the same site. In other words, different cell lineages or sublineages may arise at different gestation stages and/or sites. To investigate this possibility, one may want to study whether the cells of different potential embryonic hematopoietic sites may contribute to different types of cells within the lymphocyte lineage. It has been shown that the differentiation potential of fetal and adult stem cells, in the liver and the bone marrow, respectively, may be different (Ikuta et al. 1990). Fourth, stem cells may arise in the embryo and
111
CD3
Fig. 5. Two-color flow cytometric analysis of cells obtained from allogeneic (BALB/c) day 8 yolk sac (A and B) and from embryonal body (C and D) reconstituted thymuses cultured for 30 days. The host thymuses used for reconstitution treated with 0.9 mM 2-deoxyguanosine for 3 days did not lead to depletion of all stem cells in the thymus (see text), so that host-derived Thyl.l"1" cells are expected. In A and B, cells were obtained from day 8 yolk-sac-cellreconstituted thymuses and stained with Thy 1.2 vs. Thyl.l and Thyl.2 vs. CD3, respectively. In C and D, cells were obtained from day 8 embryonal-body- cellreconstituted thymuses and stained with Thyl.2 vs. Thyl.l and Thyl.2 vs. CD3, respectively.
migrate to the newly formed yolk sac prior to or at day 7 to 8 when the yolk sac is formed. Although the 8-day mouse embryo is at a developmental stage earner or equivalent to that used in the chick-quail embryo studies, the possibility that stem cells within the embryonic shield migrate out by day 7 of embryogenesis cannot be excluded. However, this possibility can be tested experimentally. It has been shown that the day 12 fetal liver but not the adult bone marrow stem cells can give rise to the cells expressing the TCR Vy3 (Ikuta et al. 1990). Our findings that the yolk sac, as early as at day 9 and probably also at day 8, can give rise to the TCR Vy3expressing cells before the formation of the liver, indicate that the yolk sac, like the liver at later stages, contains stem cells for at least the first wave of cells appearing in the fetal thymus. It also raises the possibility that the stem cells in fetal liver are derived from the yolk sac of earlier stages. To study this, one needs to have suitable markers so that the cell flow can be traced. An approach using microinjection of genetically marked cells into the blastocyst is currently under way. During thymus ontogeny, TCR Vy3+ cells are not detectable in the fetal thymus after day 18, although the presence of Vy3 transcripts can be detected using the polymerase chain reaction method (Havran et al. 1989; Havran and Allison, 1988; Aguilar and Belmont, 1991). It is likely that these cells have migrated from the
1322
C.-P. Liu and R. Auerbach
thymus to the skin where they form the resident Thyl + dendritic epithelial cells (Havran and Allison, 1988, 1990; Havran et al. 1989; Asarnow et al. 1988). In reconstituted thymuses, on the other hand, the TCR Vy3 + cells are present even after prolonged time in culture (Table 1 and 2) suggesting that such cells may persist if they do not migrate out of the thymus. We have been able to detect donor yolk-sac-derived cells in the culture supernatant (data not shown), which suggests that at least some thymic cells can migrate out of the reconstituted thymus in the culture system. This in vitro exit of lymphocytes may, however, be inefficient when compared to the events occurring in vivo. Despite the continued presence of TCR Vy3 + cells, the percentage of these cells was always less than 60 % of the CD3 + cells present in day 9 yolk-sac-reconstituted thymuses. The level of this percentage is similar to that in later stage yolk sac (day 10 to 13) or fetal liver (day 12 or 14)-stem-cell-reconstituted thymuses (Ikuta et al. 1990), but is lower than that (more than 90% of CD3 + cells) present in freshly obtained, non-cultured day 14 fetal thymuses as shown in a previous report (Havran et al. 1989). This suggests that cells expressing TCR other than the ones that appear among the first wave of cells in non-treated thymus are present in preliver stage yolk sac, and later stage yolk sac and fetal liver-stem-cell-reconstituted thymuses. It is possible that 2-deoxyguanosine treatment may bring in some artifact. For example, it has been suggested that bonemarrow-derived non-epithelial cells participate in the positive and negative selection processes in the thymus (reviewed in Blackman et al. 1990; Ramsdell and Fowlkes, 1990; Sprent et al. 1990). The 2-deoxyguanosine treatment kills all the non-epithelial cells but not the epithelial cells in the day 14 fetal thymus; thus the.thymic selection processes may not be carried out properly as compared to that in the normal non-treated day 14 thymuses. Therefore, even in reconstituted thymuses cultured for a shorter time, e.g. 9 days, the percentage of the TCR Vy3+ cells is lower than that present in the non-treated thymuses. Although our finding that CD4/CD8 single-positive and CD3-positive cells can be derived from day 9 yolk sac cells suggests that they may be functional cells, we have not tested their functional capacity directly due to the small number of available reconstituted thymuses and the low cell yield from day 9 yolk-sac-cellreconstituted thymus lobes (Fig. 1). On the other hand, we have previously carried out functional tests using 11day yolk-sac-reconstituted thymuses (Liu and Auerbach, 1991). In these studies, yolk-sac-derived cells were able to give proliferative responses to both antiTCR a/f5 and to anti-CD3, indicating that phenotypic changes correlate with functional reactivity. Our studies have focussed on the potential of yolk sac cells to differentiate into T-lineage cells when permitted to develop within a thymic organ microenvironment, and we have shown that by day 8 of development stem cells with the potential for T-cell differentiation can be found in the yolk sac. As we learn more about the conditions under which differentiation can be achieved
in vitro, we hope to be able to resolve the question of the initial site(s) within the embryo or extraembryonic membranes at which a common pluripotent hematopoietic stem cell can first be identified. We want to thank L. Kubai, J. Bielich and W. Auerbach for their helpful assistance, and J. Ewart and A. Globerson for valuable discussions. We also appreciate the reagents generously provided by Dr J. P. Allison (the anti-TCR Vy3 mAb, 536), Dr J. A. Bluestone (the anti-CD3e mAb, 145-2C11), and Dr M. J. Bevan (the anti-TCR V/38 mAb, F23.1). This work was supported by grants from the National Science Foundation and the National Institutes of Health.
References AGUTLAR, L. K. AND BELMONT, J. W. (1991). Vy3 T cell receptor rearrangement and expression in the adult thymus. J. Immun. 146, 1348-1352. ASARNOW, D. M., KUZIEL, W. A., BONYHADI, M., TIGELAAR, R.
E., TUCKER, P. W. AND ALLISON, J. P. (1988). Limited diversity of yd antigen receptor genes of Thy-1 + dendritic epidermal cells. Cell 55, 837-847. AUERBACH, R. (1976). Ontogeny of immune responsiveness in the mouse. Ann. Immunol, (inst. Past.) IT!, 983-992. AUERBACH, R., CAIRNS, J. S. AND ROEHM, N. (1981). Development
of the mouse immune system. Fortschr. Zoo). 26, 287-305. BLACKMAN, M., KAPPLER, J. AND MARRACK, P. (1990). The role of
the T cell receptor in positive and negative selection of developing T cells. Science 248, 1335-1341. CEREDIG, R. (1988). Differentiation potential of 14-day fetal mouse thymocytes in organ culture. Analysis of CD4/CD8-defined single-positive and double-negative cells. J. Immun. 141, 355-362. CEREDIG, R., MACDONALD, H. R. AND JENKJNSON, E. J. (1983).
Flow microfluorometnc analysis of mouse thymus development in vivo and in vitro. Eur. J. Immun. 13, 185-190. DIETERLEN-LIEVRE, F. (1987). Haemopoietic cell progenitors in the avian embryo: origin and migrations. Ann. N.Y. Acad. Sci. 511, 77-87. EREN, R., AUERBACH, R. AND GLOBERSON, A. (1987). T cell
ontogeny: extrathymic and intrathymic development of embryonic lymphohematopoietic stem cells. Immunol. Res. 6, 279-287. EREN, R., ZHARHARY, D . , ABEL, L. AND GLOBERSON, A. (1987).
Ontogeny of T cells: development of pre-T cells from fetal liver and yolk sac in the thymus microenvironment. Cell Immunol. 107, 76-84. EWART, J. L. AND AUERBACH, R. (1989). Trisomy 16 mice show decreased expression of thymocyte differentiation markers in late fetal development. Cell Differ. Develop. 27, S93. GOODMAN, T. AND LEFRANCOIS, L. (1989). Intraepithelial
lymphocytes. Anatomical site, not T cell receptor form, dictates phenotype and function. / . exp. Med. 170, 1569-1581. HAVRAN, W. L. AND ALLISON, J. P. (1988). Developmental^
ordered appearance of thymocytes expressing different T-cell antigen receptors. Nature 335, 443-445. HAVRAN, W. L., GRELL, S., D U W E , G., KIMURA, J., WILSON, A., KRUISBEEK, A. M., O'BRIEN, R. L., BORN, W., TIGELAAR, R. E.
AND ALUSON, J. P. (1989). Limited diversity of T-cell receptor gamma-chain expression of murine Thy-1 + dendritic epidermal cells revealed by V 3-specific monoclonal antibody. Proc. natn. Acad. Sci. U.S.A. 86, 4185-4189. HAVRAN, W. L. AND ALLISON, J. P. (1990). Origin of Thy-1 +
dendritic epidermal cells of adult mice from fetal thymic precursors. Nature 344, 68-70. HUSMANN, L. A., SHIMONKEVTTZ, R. P., CRISPE, I. N. AND BEVAN,
M. J. (1988). Thymocyte subpopulations during early fetal development in the BALB/c mouse. J. Immun. 141, 736-740. IKUTA, K., KINA, T., MACNEIL, I., UCHIDA, N., PEAULT, B.,
CHIEN, Y.-H. AND WEISSMAN I. L. (1990).. A developmental
Yolk sac origin of murine T cells 1323 switch in thymic lymphocyte maturation potential occurs at the level of hematopoietic stem cells. Cell 62, 863-874. JENKINSON, E. J., FRANCHI, L. L., KINGSTON, R. AND OWEN, J. J.
T. (1982). Effect of deoxyguanosine on lymphopoiesis in the developing thymus rudiment in vitro:application in the production of chimeric thymus rudiments. Eur. J. Immun. 12, 583-587. KlSIELOW, P . , LEISERSON, W . AND VON BOEHMER, H . (1984).
Differentiation of thymocytes in fetal organ culture: analysis of phenotypic changes accompanying the appearance of cytolytic and interleukin 2-producing cells. /. Immun. 133, 1117-1121. KUBAI, L. AND AUERBACH, R. (1983). A new source of embryonic lymphocytes in the mouse. Nature 301, 154-156. KUBO, R. T., BORN, W., KAPPLER, J. W., MARRACK, P. AND
PIGEON, M. (1989). Characterization of a monoclonal antibody which detects all murine afi T cell receptors. /. Immun. 142, 2736-2742. LEO, O., FOO, M., SACHS, D. H., SAMELSON, L. E. AND
BLUESTONE, J. A. (1987). Identification of a monoclonal antibody specific for a munne T3 polypeptide. Proc. natn. Acad. Sci. U.S.A. 84, 1374-1379. Liu, C.-P. AND AUERBACH, R. (1991). Ontogeny of munne T cells: thymus-regulated development of T cell receptor-bearing cells derived from embryonic yolk sac. Eur. J. Immun. 21, 1849-1855. METCALF, D. AND MOORE, M. A. S. (1971). Haemopoietic Cells. Amsterdam: North Holland Publishing Co. MOORE, M. A. S. AND OWEN, J. J. T. (1967). Experimental studies on the development of the thymus. /. exp. Med. 126, 715-725.
RAMSDELL, F. AND FOWLKES, B. J. (1990). Clonal deletion versus clonal anergy: the role of the thymus in inducing self tolerance. Science 248, 1342-1348. RUCH, R. (1968). The Mouse: Its Reproduction and Development. Burgess Publishing Co., MN. SPANGRUDE, G. J., HEIMFELD, S. AND WEISSMAN, I. L. (1988).
Purification and characterization of mouse hematopoietic stem cells. Science 241, 58-62. SPRENT, J., GAO, E.-K. AND WEBB, S. R. (1990). T cell reactivity to MHC molecules: Immunity versus tolerance. Science 248, 1357-1363. STAERZ, U. D., RAMMENSEE, H., BENEDETTO, J. D. AND BEVAN,
M. J. (1985). Characterization of a murine monoclonal antibody specific for an allotypic determinant on T cell antigen receptor. J. Immun. 134, 3994-4000. TOLES, J. F., CHUI, D. H. K., BELBECK, L. W., STARR, E. AND
BARKER, J. E. (1989). Hematopoietic stem cells in murine embryonic yolk sac and peripheral blood. Proc. natn. Acad. Set. U.S.A. 86, 7456-7459. WEISSMAN, I. L., PAPAIOANNOU, V. E. AND GARDNER, R. L.
(1978). Fetal hematopoietic origins of the adult hematolymphoid system. In Differentiation of Normal and Neoplastic Hematopoietic Cells (ed. B. Clarkson, P. A. Marks and J. E. Till), pp. 33-47, Cold Spring Harber, New York: Cold Spring Harbor Laboratory. Wu, L., SCOLLAY, R., EGERTON, M., PEAUSE, M., SPANGRUDE, G.
J. AND SHORTMAN, K. (1991). CD4 expressed on earliest Tlineage precursor cells in the adult murine thymus. Nature 349, 71-74.
OWEN, J. J. T. AND JENKINSON, E. J. (1981). Embryology of the
lymphoid system. Prog. Allergy 29, 1-34.
{Accepted 19 August 1991)
