|
|
||||||||
BRIEF REVIEWS |

* Department of Cell Biology and Histology, Academic Medical Center of the University of Amsterdam, Amsterdam, The Netherlands; and
Netherlands Cancer InstituteAntoni van Leeuwenhoek Ziekenhuis, Amsterdam, The Netherlands
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The discovery of the scid mutation in the C.B-17 mouse strain (2) opened the door to extensive research for the design of such animal models. These C.B-17 mice harbor a mutation of the prkdc gene involved in rearrangement of TCR and Ig genes and therefore mostly lack mature T and B lymphocytes (3). The first report of xenogeneic transplantation of human PBL in C.B-17 SCID mice was published in 1988, and only the transferred mature T and B cells survived during a limited period of time in this human (hu) 2-PBL-SCID model (4). Obviously, the presence of human pluripotent progenitor cells is required to ensure development of multiple hemopoietic lineages. Therefore, human fetal thymic lobes and pieces of fetal liver (FL) were transplanted under the kidney capsule in C.B-17 SCID mice (5). In this difficult experimental setting, the developing human PBL found were almost exclusively T cells. These cells migrated from the human thymus graft to the periphery after 34 wk. However, only one-third to one-half of the animals had a detectable population of human cells in the blood (0.7% huCD45+ cells on average in the lymphocyte gate) 35 mo posttransplant (6). This SCID-hu (thymus (Thy)/liver (Liv)) model has been extensively used for the analysis of human hemopoiesis, especially T cells, and as an animal model of HIV infection (1).
In another approach, sublethally irradiated C.B-17 SCID mice were repopulated with progenitor cells from human bone marrow (BM) or umbilical cord blood (UCB). Such progenitors contain cells with SCID repopulating ability (SRC), and this system was therefore referred as to hu-SRC-SCID model (7, 8). Similarly, human myeloid differentiation was obtained after transfer of human total BM cells into athymic NK-deficient beige/nude/xid SCID mice (9). Although UCB cells appeared to be more efficient, only low efficiencies and levels of engraftment were achieved. Furthermore, T cell development was always extremely limited, supporting the original observation of a virtual absence of human cells in the murine thymus of SCID-hu (Thy/Liv) mice (5).
The genetic background of the recipient animals appears to determine the success of engraftment (for review, see Ref.10). For example, NOD/SCID mice were a more receptive strain for human cell engraftment than the C.B-17 SCID mice (11, 12). Mice in the NOD genetic background exhibit deficiencies in 1) NK cell activity, at least partially due to impairment of the activating receptor NKG2D (13); 2) complement activation, due to C5 deficiency (14); and 3) LPS-induced production of IL-1 by macrophages (15). Unfortunately, NOD/SCID mice show an elevated tendency to develop thymic lymphomas, hindering the feasibility of long-term studies (16). When NOD/SCID mice were engrafted with CD34+ cells from human UCB or adult BM, human T cell progenitors could be identified and isolated from the BM of the animals, and these cells gave rise to T cells in fetal thymic organ cultures (17). Although T cell development occurred only rarely in NOD/SCID mice, ablation of NK cells by using the CD122/IL-2R
-specific Ab TM-
1 significantly increased the frequency of thymopoiesis in animals reconstituted with human UCB cells (18). Since the host murine NK cells appeared to be particularly limiting, human T cell development was also assessed in NOD/SCID/
c/ mice (19) and NOD/SCID/
2m/ mice (20). Both
2m and
c deficiencies result in reduction of the activities and numbers of NK cells (21, 22). However, although human thymopoiesis in situ was significantly increased, as compared with experiments using NOD/SCID mice, T cell repopulation of the periphery remained limited (18, 19, 20).
In summary, there were several factors that limited the impact of these models in these earlier studies: 1) the engraftment of human self-renewing hemopoietic progenitors was highly variable; 2) differentiation of the injected human progenitor cells rarely gave rise to all hemopoietic-derived lineages; 3) intrinsic limitations of the murine recipients are problematic; and 4) in situ (i.e., in murine thymus) development of human T cells, if any, remained suboptimal.
| Identification of human progenitors that repopulate SCID mice and influence of cytokines |
|---|
|
|
|---|
2m/,
c/) or NK-depleted (anti-IL-2R
) NOD/SCID mice (28, 29, 30). In contrast, long-term repopulating cells transplanted into NOD/SCID mice do not disappear after 68 wk and develop predominantly into highly dividing B cells with a poor capacity of terminal differentiation. Indeed, the BM of such mice contains mostly CD34+CD19+ pro-B cells and the spleen CD19+IgM+ immature B cells unable to undergo class switch recombination (31, 32, 33). The use of SCID models in combination with IL treatments also gave insight into the requirements for xenogeneic engraftment of human progenitors. HuIL-3 treatment, but not IL-2, IL-7, Flt-3 ligand, nor stem cell factor (SCF; kitL) alone, was shown to enhance the engraftment by human BM progenitors in beige/nude/xid mice from <1% huCD45+ cells in the blood up to 4% (9, 34). Similarly, cotreatment with erythropoietin, SCF, and/or GM-CSF/IL-3 fusion protein in BM-transplanted C.B-17 SCID mice enhances the proportion of human cells from <1% to 1015% in treated animals, with development of multiple hematological cell lineages (7). Such a treatment was ineffective when UCB cells were used instead of BM cells, since UCB already showed a high engraftment capacity (15% huCD45+ cells in BM) in the absence of cytokines (8). Overall, the outcome of cells with SCID repopulating capacity appears to be highly heterogeneous, depending on the SCID model used, and supplementation by human cytokines may improve engraftment in these models.
The human immune system (HIS) Rag2/ c/ model
|
|---|
|
|
|---|
c/ mice appeared to be of major interest because they lack T, B, and NK cells (36).
Initial work done by reconstituting adult Rag2/
c/ mice with human stem cells from cord blood or FL was proven not successful as far as T cell development was concerned. Similar poor thymopoiesis (6 vs 18% animals with active human T cell development) was observed in BALB/c Rag2/
c/ and NOD/SCID mice (18, 37). Treatment of CD34+ UCB cell- reconstituted C57BL/6 Rag2/
c/ mice with a mixture of IL-3, GM-CSF, and erythropoietin showed limited improvement of human engraftment (38). These observations suggested that the BM environment in Rag2/
c/ mice lacks the capacity to produce growth factors necessary for optimal engraftment. A not mutually exclusive possibility is that the murine macrophages hinder the engraftment, because of noxious effects on human progenitors and their mature progeny, or niche occupancy.
Recently, Manz and coworkers (39) and our group (40) have administered human HSC to sublethally irradiated BALB/c Rag2/
c/ (BALB-Rag/
) newborn mice (Fig. 1). This approach has been proven successful and reliable, since >80% of the obtained HIS mice exhibit 10% or more human CD45+ cells in peripheral blood and other lymphoid organs. Animals reconstituted with CD34+ UCB cells have been referred as to "human adaptative immune system Rag2/
c/ mice" (huAIS-RG) by the group of Manz and colleagues (41). All major immune cell subsets are produced de novo in the HIS (BALB-Rag/
) mice within 6 wk, e.g., T, B, NK cells, monocytes, conventional, and plasmacytoid dendritic cells. The efficiency of engraftment by i.p. injected HSC is extremely age dependent, since 1-day-old animals give rise to
80% human cells in the blood 8 wk after reconstitution, whereas only
30% human cells are obtained with 1-wk-old animals and <10% with 2-wk-old mice (40). We compared the effect of the route of injection of CD34+ FL cells, namely, i.p (40) vs intrahepatic (i.h.) (39). When CD34+ FL cells were inoculated i.h., engraftment was more reliable, but the effective level of human chimerism was not different between i.p. and i.h. routes (our unpublished observations). Routinely, the percentage of human cells in HIS (BALB-Rag/
) mice 8 wk after reconstitution reaches >95% in thymus, >60% in blood and liver, and >50% in spleen and BM, although variability is observed. This variability is not due to the source of the CD34+ cells, namely, from 14 to 17 wk of gestation FL (40) or from UCB (39), but rather directly linked to the number of injected progenitor cells. As compared with the previously available models (Table I), reconstitution of newborn mice represented a major improvement, and this was also recently observed with NOD/SCID/
c/ mice (which show no signs of lymphoma development) and, to a lesser extent, NOD/SCID/
2m/ mice (42, 43).
|
|
c/ mice, resulting in thymic rudiments unable to support human T cell development. 2) Another possibility is that phagocytic cells present in newborn mice, e.g., macrophages or neutrophils, are less limiting than those in adults, resulting in less resistance in newborns. The latter possibility is supported by observations showing efficient reconstitution in adult BALB/c Rag2/
c/ mice with FL CD34+ cells (our unpublished observations) or CD34+ UCB cells (44), only if the recipient mice were previously treated with clodronate (Cl2MBP; dichloromethylene-bisphosphonate)-containing liposomes. These liposomes can be used as a tool to deliver drugs to phagocytic cells, especially macrophages, and the accumulation of clodronate leads to irreversible metabolic damages, which will eventually result in apoptosis (45). Treatment of adult BALB/c Rag2/
c/ mice with clodronate-containing liposomes was already shown to improve the speed and efficiency of engraftment by human PBL (46). These observations indicate that phagocytic cells, presumably macrophages, are a particularly limiting factor for xenogeneic engraftment in adult Rag2/
c/ mice. This could explain the positive results previously obtained in NOD/SCID/
c/ and NOD/SCID/
2m/ mice, which combine both myeloid defect of the NOD background and NK cell deficiency (19, 20). | Influence of murine genetic background on repopulation efficiency |
|---|
|
|
|---|
c/ animals as recipients for the human HSC (39, 40). Similarly, the original C.B-17 SCID mice were made in a BALB/c background (5). As already mentioned, the impact of the scid mutation on human cell engraftment was tested in different genetic backgrounds (11), including the C3H strain with defective macrophage response to LPS, the complement C5-deficient DBA/2 strain and the C57BL/6J strain, in which NK cells can be depleted by anti-NK1.1 Ab treatment (10). Only NOD/SCID mice combining complement, macrophage, and NK deficiencies showed an improved capacity for xeno-engraftment. It is likely that the presence of fully competent complement, NK cells, and/or macrophages in the other tested SCID mouse strains limited such a comparison (11). The use of "nonleaky" Rag2/
c/ mice could therefore give information about the relative strength of the innate immune system, in relation to the genetic background, and the direct comparison of the two existing strains, i.e., BALB/c vs C57BL/6 Rag2/
c/ mice, will be extremely informative. Interestingly, differences were already noticeable after reconstitution of adult animals with human UCB cells: analysis of the BM showed that engraftment in the C57BL/6 background was extremely poor (<1% CD45+ cells), whereas NOD/SCID and BALB/c Rag2/
c/ mice appeared to be similar (around 20% chimerism) (18, 38). | T cell development in HIS mice |
|---|
|
|
|---|
) mice (39, 40). The murine thymus of HIS (BALB-Rag/
) mice supports human T
and T
cell development. This gives rise to the establishment of cortex- and medulla-like regions within the thymic lobules of the mouse thymus. At the age of 8 wk, the proportion of double-positive CD4+CD8+ thymocytes ranges from 50 to 80%, indicating that sustained thymopoiesis occurs in HIS (BALB-Rag/
) mice. This percentage decreases in an age-dependent fashion, but a large proportion of CD4+CD8+ thymocytes can still be observed 4- 6 mo after inoculation of CD34+ FL cells. In young HIS (BALB-Rag/
) mice (46 wk after injection of FL CD34+ cells), the number of human thymocytes ranges from 5 x 106 to 15 x 106 cells, as compared with 210 x 106 cells in young adults (812 wk after inoculation) and 0.52 x 106 cells in old animals (>4 mo after inoculation). Mature 
T cells are 2030 times more abundant than 
T cells. Furthermore, human CD16+CD56+ NK cells are also observed in the murine thymus.
Human mature T cells start seeding the peripheral lymphoid organs around 45 wk after reconstitution. They populate the BM (
1% of CD45+ human cells), spleen, liver, lungs, and peripheral blood (520%), with a CD4:CD8 ratio of 3:14:1. These T cells proliferate after in vitro stimulation with allogeneic human dendritic cells (39) or irradiated PBL in presence of PHA and huIL-2 (our unpublished observations). Furthermore, the number of mature T cells can be transiently enhanced in vivo by treatment with a superagonist anti-CD28 Ab (N. Legrand, T. Cupedo, A. U. van Lent, M. Ebeli, K. Weijer, T. Hanke, and H. Spits, submitted for publication), which has been described for its unique property of inducing human T cell activation independently of TCR stimulation (47). These observations indicate that human T cells developing in HIS (BALB-Rag/
) mice are functional. Interestingly, a CD4+CD25+ glucocorticoid-induced TNFR family-related gene subset of T cells, enriched for foxp3 expression, is also observed both in the thymus and in the periphery (39) (N. Legrand, T. Cupedo, A. U. van Lent, M. Ebeli, K. Weijer, T. Hanke, and H. Spits, submitted for publication), strongly suggesting that they represent human T cells exerting regulatory functions (48).
| Selecting MHC molecules of human T cells in HIS mice |
|---|
|
|
|---|
) mice implies that these cells undergo a process of positive selection. Our current findings with BALB/c nude-Rag2/
c/ mice, which do not show any T cells after engraftment of CD34+ FL cells, strongly indicate that no thymus-independent T cell development is achieved in the HIS mice (our unpublished observations). In theory, developing human T cells in HIS (BALB-Rag/
) mice could be positively selected both by thymic epithelial cells of the recipient mouse and by BM-derived human cells, especially in suboptimal conditions (49). Results obtained so far indicate that this dual selection could indeed occur, since MLRs show no/low T cell proliferation against BALB/c-derived or autologous human dendritic cells, i.e., tolerance/negative selection is achieved (39). Still, the absence of a reaction does not necessarily indicate that the circulating T cells in HIS mice are restricted to mouse or human MHC molecules without any bias, although the generated T cell repertoire is broad and indistinguishable from the repertoire of a normal donor (39, 40). For instance, after infection of HIS (BALB-Rag/
) mice with influenza, we only observed T cells specific for viral epitopes in the context of murine MHC molecules (our unpublished observations). This result could reflect a biased T cell repertoire to mouse MHC molecules in the periphery, although thymic selection is performed practically in virtue of TCR affinities for both human and murine MHC molecules. Alternatively, the nature of the targeted tissues (mouse vs human) by the used infectious agents could result in the specific activation of T cell repertoires that were selected by mouse or human MHC molecules. In support of this argument, EBV infection of HIS (BALB-Rag/
) mice was shown to give rise to human T cells that proliferate ex vivo when cultured on EBV-transformed B cells (39). This observation suggests that the generated T cells are specific for EBV epitopes presented by human B cells in the context of human MHC molecules.
In addition, it was previously reported that the half-life of T cells in the classical SCID-hu (Thy/Liv) model is <24 h (50). This observation could be due to the lack of MHC elements supporting the peripheral survival of T cells selected by human MHC molecules in the thymic graft. Interestingly, we observed that human T cells in HIS (BALB-Rag/
) mice also display a high turnover rate and are prone to apoptosis (N. Legrand, et al., submitted for publication), despite the fact that they are potentially selected by both human and murine MHC molecules. This observation could reflect a mismatch between central selection and peripheral survival requirements of these T cells, negative effects due to the murine innate immune system, and/or lack of factors involved in T cell homeostasis.
| Development of other hemopoietic lineages in HIS mice |
|---|
|
|
|---|
) mice. The murine BM supports human B cell differentiation and contains mainly immature IgMCD10+CD20 B cells, whereas most B cells in the spleen were IgM+IgD+CD10+CD20+ (39, 40). Human IgM and IgG produced by CD19+CD27+CD138+ plasma cells are gradually accumulating with time in the serum, demonstrating complete B cell maturation in this system, in contrast to previous results in NOD/SCID mice (32). HIS (BALB-Rag/
) mice vaccinated with tetanus toxoid 1217 wk after transplantation of CD34+ UCB cells could develop a specific IgG response, and memory B cells, although, for yet to be understood reasons, this was not the case if animals received tetanus toxoid only 8 wk after inoculation of human progenitors (39). Furthermore, the presence of B cells induces de novo generation of mouse follicular dendritic cells, presumably in a human lymphotoxin-
-dependent manner, and formation of germinal center-like structures (39).
Other important lineages develop from the human progenitors in the HIS (BALB-Rag/
) mice. We reported low frequencies of CD56+ NK cells in all analyzed organs, the thymus being the organ where most NK cells are observed (40). Similarly, CD14+ monocytes and CD11c+ conventional dendritic cells, which can be used ex vivo as stimulators of allogeneic T cells, were also observed in low amounts, except in the BM and liver (39, 40). CD123/IL-3R
+BDCA2+CD11c plasmacytoid dendritic cells, also referred to as natural IFN-producing cells (51), are developing in HIS (BALB-Rag/
) mice and accumulate in high frequencies in the BM and the liver. Upon ex vivo stimulation with influenza virus, these plasmacytoid dendritic cells produce high levels of IFN-
(39). Furthermore, the HIS (BALB-Rag/
) mouse model has been used in our laboratory to demonstrate that the expression of the Spi-B transcription factor is needed for the correct development of human plasmacytoid dendritic cells by using a small interfering RNA "knock-down" approach (52).
Overall, the HIS (BALB-Rag/
) mouse model supports the development of both the human innate and adaptive immune systems, and it can be manipulated to answer questions linked to developmental biology (40, 52) or treatment purposes.
| Future prospects |
|---|
|
|
|---|
Nonetheless, it is clear that Rag2/
c/ mice are not the most optimal recipients to achieve fully operational human immunity. As a result of the disruption of the IL-7/IL-7R signaling axis in these animals, functionality of CD3CD4+IL-7R
+ lymphoid tissue inducer cells is compromised and formation of most lymph nodes is therefore impaired (53). Consequently, the search for a mouse strain lacking NK cells, but still harboring normal lymph node development is of interest. Rag2/IL-2R
/ mice are good candidates since these mice lack T, B, and NK cells (54) but, in contrast to Rag2/
c/ mice, this strain is expected to develop normal lymph nodes. Furthermore, based on the original observations of improved engraftment after infusion of cytokine mixtures (7, 9, 34, 38) and the relative bias toward B cell differentiation in several SCID mouse models, it is suggested that murine GM-CSF, M-CSF, IL-3, and thrombopoietin (which support myeloid/erythroid differentiation) are not cross-reactive with human cells, whereas Flt3 ligand, IL-7 (which support lymphoid differentiation), and SCF are cross-reactive, at least to some extent. It may therefore be of interest to insert the genes encoding human IL-3, TPO, SCF, and/or GM-CSF into the host mice, as well as other cytokines, like IL-15 for improvement of NK cell development. As an alternative for this gene knock-in approach, infusion of molecules with increased stability can be used. Using a fusion protein between the Fc fragment of human Ig and IL-7, it was recently shown that the proportion of human cells in NOD/SCID/
c/ mice engrafted with CD34+ cells from human mobilized PBLs increased by a factor of 23 in thymus and blood (55). Molecules involved in the development of the innate immune system, e.g., TLRs, could also be targeted to improve its establishment in the mouse or to help for an optimal initiation of immune responses. Finally, by introducing human class I (HLA-A/B/C) and class II (HLA-DP/DQ/DR) molecules of the MHC, both central selection of human T cells and their peripheral survival capacity will certainly be improved. Several groups are currently working on these extremely challenging approaches, and exiting results should arise in the coming years. It is highly likely that these efforts will provide accurate models for in vivo preclinical studies for human infectious diseases and genetic disorders.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 Address correspondence and reprint requests to Dr. Hergen Spits, Department of Cell Biology and Histology, Academic Medical Center, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. E-mail address: hergen.spits{at}amc.uva.nl ![]()
2 Abbreviations used in this paper: hu, human; FL, fetal liver; BM, bone marrow; HIS, human immune system; HSC, hemopoietic stem cell; i.h., intrahepatic; SRC, SCID repopulating cell; UCB, umbilical cord blood;
c, common
-chain;
2m,
2-microglobulin; SCF, stem cell factor; Thy, thymus; Liv, liver. ![]()
Received for publication October 21, 2005. Accepted for publication November 21, 2005.
| References |
|---|
|
|
|---|
cnull mice model. Blood 102: 873-880.
2 microglobulin-deficient (B2m(null)) NOD/SCID mice are excellent recipients for studying human stem cell function. Blood 95: 3102-3105.
chain. Proc. Natl. Acad. Sci. USA 92: 377-381.
2 microglobulin-null mice. J. Clin. Invest. 107: 199-206. [Medline]
chain double mutantsrequiring exogenous cytokine administration for human hematopoietic stem cell engraftment. J. Interferon Cytokine Res. 19: 533-541. [Medline]
c/ mice: functional inactivation of p53 in developing T cells. Blood 104: 3886-3893.
c/ mice. Ann. NY Acad. Sci. 1044: 236-243.
chain null mice. Blood 106: 1565-1573.
c double-knockout mice after treatment with CL2MDP liposomes. Exp. Hematol. 32: 1118-1125. [Medline]
c/ double-mutant mice. Blood 102: 2522-2531.
. Science 268: 1472-1476.
null mice engrafted with mobilized human hemopoietic stem cells. J. Immunol. 174: 6477-6489.
c(null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood 100: 3175-3182.
null mice. J. Immunol. 169: 204-209. This article has been cited by other articles:
![]() |
J. Y. Kausman and A. R. Kitching A New Approach to Idiopathic Nephrotic Syndrome J. Am. Soc. Nephrol., October 1, 2007; 18(10): 2621 - 2622. [Full Text] [PDF] |
||||
![]() |
D. S. An, B. Poon, R. H. T. Fang, K. Weijer, B. Blom, H. Spits, I. S. Y. Chen, and C. H. Uittenbogaart Use of a Novel Chimeric Mouse Model with a Functionally Active Human Immune System To Study Human Immunodeficiency Virus Type 1 Infection Clin. Vaccine Immunol., April 1, 2007; 14(4): 391 - 396. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. D. SHULTZ, T. PEARSON, M. KING, L. GIASSI, L. CARNEY, B. GOTT, B. LYONS, A. A. ROSSINI, and D. L. GREINER Humanized NOD/LtSz-scid IL2 Receptor Common Gamma Chain Knockout Mice in Diabetes Research Ann. N.Y. Acad. Sci., April 1, 2007; 1103(1): 77 - 89. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. KING, T. PEARSON, L. D. SHULTZ, J. LEIF, R. BOTTINO, M. TRUCCO, M. ATKINSON, C. WASSERFALL, K. HEROLD, J. P. MORDES, et al. Development of New-Generation HU-PBMC-NOD/SCID Mice to Study Human Islet Alloreactivity Ann. N.Y. Acad. Sci., April 1, 2007; 1103(1): 90 - 93. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Gorantla, H. Sneller, L. Walters, J. G. Sharp, S. J. Pirruccello, J. T. West, C. Wood, S. Dewhurst, H. E. Gendelman, and L. Poluektova Human Immunodeficiency Virus Type 1 Pathobiology Studied in Humanized BALB/c-Rag2-/-{gamma}c-/- Mice J. Virol., March 15, 2007; 81(6): 2700 - 2712. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Talmadge, R. K. Singh, I. J. Fidler, and A. Raz Murine Models to Evaluate Novel and Conventional Therapeutic Strategies for Cancer Am. J. Pathol., March 1, 2007; 170(3): 793 - 804. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. B. de la Grange, F. Armstrong, V. Duval, M.-C. Rouyez, N. Goardon, P.-H. Romeo, and F. Pflumio Low SCL/TAL1 expression reveals its major role in adult hematopoietic myeloid progenitors and stem cells Blood, November 1, 2006; 108(9): 2998 - 3004. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Baenziger, R. Tussiwand, E. Schlaepfer, L. Mazzucchelli, M. Heikenwalder, M. O. Kurrer, S. Behnke, J. Frey, A. Oxenius, H. Joller, et al. Disseminated and sustained HIV infection in CD34+ cord blood cell-transplanted Rag2-/-{gamma}c-/- mice PNAS, October 24, 2006; 103(43): 15951 - 15956. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Legrand, T. Cupedo, A. U. van Lent, M. J. Ebeli, K. Weijer, T. Hanke, and H. Spits Transient accumulation of human mature thymocytes and regulatory T cells with CD28 superagonist in "human immune system" Rag2-/-{gamma}c-/- mice Blood, July 1, 2006; 108(1): 238 - 245. [Abstract] [Full Text] [PDF] |
||||
| ||||||