|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on June 21, 2002; DOI 10.1182/blood-2002-04-1036.
IMMUNOBIOLOGY
From the Departments of Biological Structure and
Immunology, Howard Hughes Medical Institute, University of Washington,
Seattle, WA; and the Laboratory of Cellular and Molecular Biology,
National Cancer Institute, National Institutes of Health, Bethesda, MD.
Here we demonstrate that keratinocyte growth factor (KGF) and
FGFR2IIIb signaling can affect development and function of thymic epithelium (TE) and that The thymus is a heterogeneous epithelial
environment where morphologically and phenotypically distinct
epithelial compartments support thymocyte development. Contributions of
thymic epithelium (TE) to this process include the elaboration of
cytokines affecting thymocyte development, the role of major
histocompatibility complex (MHC)-peptide complexes expressed
by cortical TE in positive selection, and the participation of
medullary TE in some models of negative selection.1
Differential expression of chemokines by TE subsets and a
developmentally regulated pattern of chemokine receptor expression by
thymocytes have been proposed as mechanisms to effect a serial exposure
of developing thymocytes to distinct epithelial compartments and may
underlie the centripetal movement of thymocytes within the
thymus.2 This association of thymocytes at different stages of maturation with distinct epithelial compartments and functional studies of transgenic and mutant mice have led to the notion
that these different epithelial compartments contribute sequentially to
the intrathymic phase of T-cell development.
There is accumulating evidence that the functional integrity and
developmental potential of thymic epithelium is not autonomous and is
dependent on signals derived from nonepithelial sources. Thymocytes
themselves contribute to the growth and differentiation of
TE,3 although the mediators responsible remain largely
undefined. Fibroblasts have also been implicated in the development of
the thymic environment. Enzymatic removal of connective tissue from fetal thymic lobes prevented their subsequent development when engrafted,4 and a requirement for mesenchyme in the
successful reconstitution of functional thymic tissue has also been
demonstrated in vitro.5 Other work indicates that this
mesenchymal contribution is not restricted to the induction of TE
differentiation but is also required for maintenance of TE
function.6
Little is known regarding the processes that underlie mesenchymal
contribution to TE growth/differentiation. Epidermal growth factor
(EGF) is a potential mesenchymally derived mediator of thymus
organogenesis,7 and fibroblast growth factor (FGF) family members are also candidate effector molecules. Members of this complex
family of polypeptides serve as ligands for cell surface receptors with
tyrosine kinase activity and have been implicated in embryonic
development and patterning.8,9 We focused on keratinocyte
growth factor (KGF, FGF-7) because this member of the FGF family
typically exhibits a paracrine mode of action, being produced by
mesenchymal cells and acting on a wide range of epithelial-derived
cells that express a unique splice variant of the FGFR2 receptor
(FGFR2IIIb).10-12 Indications that KGF or other FGF family
members that signal through this receptor could play a role in thymic
organogenesis include the in situ demonstration of KGF message in the
fetal thymus10 and thymic dysgenesis in transgenic mice
expressing soluble dominant-negative FGFR2IIIb receptor.13
Furthermore, mice lacking either FGF-10 or the IIIb form of FGFR2
display hypoplastic thymic tissue.14 Here we show that
Reagents
Animals and cell preparation
Fetal thymic organ culture Fetal thymic organ culture (FTOC) was performed as described19 with the filter membranes supported on stainless steel screens in 2.5 mL of media. Culture in deoxyguanosine (DOG) to deplete hematogenous cells was done as described.20 DOG-treated lobes were dissociated at 37°C with a mixture of dispase (0.8 U/mL), collagenase (0.1 U/mL), and DNAse (150 U/m) (Boehringer Mannheim) in Hanks balanced salt solution (HBSS) for biochemical and flow cytometric analyses or homogenized in TRIzol (Gibco BRL, Grand Island, NY) for subsequent reverse transcription-polymerase chain reaction (RT-PCR) analyses.Flow cytometry Fetal thymic lobes were mechanically dispersed in cold medium, passed through nylon mesh, and washed with additional medium. To minimize Fc receptor-mediated labeling, cells were incubated with anti-Fc RII Mab 2.4G2 (42; 50% hybridoma supernatant in HBSS supplemented with 0.1% NaN3), 1% fetal bovine serum, 10%
rat serum, and 10% goat serum prior to labeling with
fluorochrome-conjugated Mabs. Cells were stained with either
anti-CD3 -FITC (clone 500A221), anti-CD4-FITC (clone
RM4-5; Pharmingen, San Jose, CA), and/or anti-CD8-PE (clone
3B5; Caltag, Burlingame, CA) and analyzed with a FACScan flow cytometer
(Becton Dickinson, Mountain View, CA). All data shown are with gates
set on living cells, as assessed by their exclusion of the fluorescent
dye 7-amino-actinomycin D (Molecular Probes, Eugene,
OR).22 Data were analyzed with CellQuest software
(Becton Dickinson).
Thymocyte fractionation Thymocytes were sorted on the basis of their CD4 and CD8 expression using a Vantage cell sorter (Becton Dickinson). CD4 and CD8 single-positive thymocytes were obtained by sequential enrichment; first depletion with biotinylated anti-CD8 monoclonal antibodies and streptavidin-conjugated magnetic beads (PerSeptive Biosystems, Farmingham, MA), then fluorescence-activated cell sorting of the recovered cells with a combination of directly labeled anti-CD4 and anti-CD8 antibodies. Triple negative (CD3, CD4, and CD8) were obtained by magnetic depletion of CD4 or CD8+ cells, followed by sorting on the basis of CD3 expression.RNA purification and cDNA synthesis Total RNA was recovered by phenol/chloroform extraction, treated with DNAse, and then quantitated spectrophotometrically. Synthesis of cDNA was performed with avian myeloblastosis virus (AMV) reverse transcriptase according to the manufacturer's recommendations (Promega, Madison, WI).Conventional PCR Sequences for the PCR primers and real-time PCR probes used are given in Table 1. Primers were purchased from Genosys (The Woodlands, TX), and fluorescent probes were purchased from Biosearch Technologies (Novato, CA). For conventional PCR, normalization was done by performing hypoxanthine-guanine phosphoribosyl transferase (HPRT) PCR analysis, separating the reaction products by electrophoresis through a 1.8% agarose gel stained with ethidium bromide and generating negative images of the gel. Negatives were scanned and processed with Image software (public domain; http://rsb.info.nih.gov/nih-image/download.html) to obtain densitometry values for individual bands. Volumes of cDNA samples were adjusted to give equivalent densitometric values. Normalization with a HPRT competitor construct23 also was employed in some studies with equivalent results.
For real-time PCR, KGF and HPRT reactions used an antibody-bound hot start Thermus aquaticus (TAQ) polymerase (Qiagen, Valencia, CA) and the same thermal cycling (50°C hold, 2 minutes; 95°C hold, 10 minutes; cycle 95°C, 20 seconds; 60°C, 1 minute). For both reactions, 50 cycles were carried out using an ABI Prism 7700 real-time thermocycler (PE Biosystems, Norwalk, CT). A comparative CT method was used to determine relative gene expression
(Table 2).24 Threshold
cycles (Ct) reflected the cycle number when the fluorescence
generated by cleavage of the fluorescent probe passed a predetermined
threshold above baseline. The threshold was adjusted above the baseline
values and to the start of the logarithmic curve of the plots, as
described by the manufacturer. There was an excellent correlation
coefficient (HPRT, 0.998; KGF, 1.000) over a 4-log dilution range, with
3.939 and 3.936 cycles representing a 10-fold change in KGF or HPRT
cDNA levels, respectively.
In situ hybridization Analyses of thymus tissue were done according to published protocols25 using digoxigenin-modified sense and antisense probes.Immunodetection of KGF in thymocyte-conditioned medium Unfractionated thymocytes were cultured at 106 cells/mL in HL-1 medium for 48 hours. After low-speed centrifugation (300g × 10 minutes), the medium was concentrated 10-fold with Centricon devices (3K cutoff, Millipore, Bedford, MA). Aliquots of the concentrated media samples were added to equal volumes of 2 × sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer, heated to 100°C for 5 minutes, and then processed for SDS-PAGE and electrophoretic transfer to nitrocellulose membranes and processed for immunoblot analyses as previously described.16 Peroxidase activity was detected with chemiluminescence (New England Nuclear, Wellesley, MA.)Immunodetection of invariant chain and cathepsin L TE cells (2 × 105/sample) were lysed on ice for 40 minutes in 20 µL of cell lysis buffer (0.5% Nonidet P-40 (Sigma), 0.15 M NaCl, 5 mM EDTA (ethylenediaminetetraacetic acid), 50 mM Tris-HCl, pH 7.2) supplemented with a protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN). Following centrifugation at 8000 rpm for 10 minutes, the lysate supernatants were normalized for protein concentration using the Bradford reagent (Pierce Chemical, Rockford, IL). Samples were boiled for 5 minutes in SDS-reducing buffer, separated by SDS-PAGE (12% acrylamide, wt/vol), and then electrophoretically transferred onto nitrocellulose membrane. Membranes were probed for invariant chain with the IN-1 Mabs26 as described.27 Cat L was detected with polyclonal rabbit antisera to mouse Cat L (a gift of A. Erickson, University of North Carolina, Chapel Hill, NC), which has been described previously.28 Affinity-purified rabbit antiactin antibody was purchased from Sigma. Binding was detected using a horseradish peroxidase-conjugated anti-rat or anti-rabbit IgG (Amersham Pharmacia Biotech, Piscataway, NJ) diluted 1:1500 and visualized by chemiluminescence.Active site labeling Equal numbers of TE cells per sample (2 × 105) were incubated for 2 hours at 37°C in the presence of 0.25 µM cysteine protease inhibitor Cbz 125I-Tyr-Ala-CN2. This radiolabeled inhibitor binds irreversibly to the active site cysteine via a thioester bond.29 Cells lysates were processed for SDS-PAGE and electrophoretic transfer as described above, and the labeled proteins visualized by exposure to Kodak BioMax MR film (Kodak, Rochester, NY).Immunohistochemistry Phenotypic analyses of cultured thymic lobes used indirect enzyme immunohistochemical procedures30 Briefly, frozen sections of tissue were serially incubated with optimal dilutions of antigen-specific or control primary antibodies. Some of the primary antibodies had been modified with N-hydroxysuccinimidyl-digoxigenin (Boehringer Mannheim) according to the manufacturer's instructions and detected with horseradish peroxidase-conjugated Fab fragments of sheep antidigoxigenin antibodies (Boehringer-Mannheim). The 3G10 Mab was detected with a peroxidase-conjugated goat anti-rat µ chain-specific antibody (Pierce Chemical). Binding of unconjugated rat IgG monoclonal antibodies to tissue sections was detected with a 3-step procedure, where unmodified primary antibodies were detected by sequential exposure to digoxigenin-conjugated goat anti-rat IgG antibodies and peroxidase-conjugated Fab fragments of goat antidigoxigenin antibodies. Peroxidase activity was revealed with 3,3'-diaminobenzidine in the presence of hydrogen peroxide.In vivo administration of KGF Recombinant KGF (500 ug/mL) dissolved in HBSS or vehicle alone was administered intraperitoneally to 4- to 6-week-old RAG2 /-deficient mice (2.5 µg/g body weight) every
other day for 9 days. This dose and administration schedule was based
on the work of Danilenko et al.31 Ten days after
initiation of treatment, thymic lobes were processed for
immunohistochemistry, flow cytometry, or obtention of thymic RNA.
Thymic epithelial stromal cells but not thymocytes express the FGFR2IIIb receptor RT-PCR analysis of RNA samples from thymi of different ages indicated that the FGFR2IIIb receptor isoform was present within the thymus at all ages examined (Figure 1A). This mRNA expression was restricted to the predominantly epithelial stromal elements, as no signal was detected in RT-cDNA samples of thymocytes from day-16 embryos (Figure 1A) or 3-week-old postnatal mice (data not shown) but was evident in samples from a thymic epithelial cell line, TE-71 (Figure 1B). Hybridization of a FGFR2IIIb RNA probe to thymic tissue sections from 4- to 6-week-old mice indicated receptor expression throughout cortical and medullary regions, with preferential labeling of the medullary regions and under the capsule, with cortical areas exhibiting lower levels of reaction (Figure 1C). A similar localization pattern was observed with an antibody that recognizes both b and c isoforms of the FGFR2 (Figure 1D).
Thymocytes are a source of KGF Analysis of cDNA samples prepared from fetal and postnatal thymi with PCR primers specific for KGF revealed the presence of KGF message (Figure 2A). The pattern of KGF mRNA expression by thymocytes at different stages of development was determined by quantitative real-time PCR of thymocyte cDNA. Analysis of cDNA prepared from thymocyte subsets sorted on the basis of CD3, CD4, and CD8 expression is shown in Table 1. KGF RT-PCR product was undetectable in the CD348
thymocyte population (with 50 cycles of PCR) and highest in the mature
thymocyte subsets, with the CD4+8+ population
showing intermediate values. Similar to the
CD348 thymocytes in normal
mice, RAG2/ thymocytes also failed to express
detectable levels of KGF signal. Results from in situ hybridization
studies were consistent with the RT-PCR analysis (Figure 2B). We
routinely observed a lightly labeled region under the capsule,
intermediate labeling throughout the cortex, and the most intense
labeling in medullary areas. That this labeling was associated with
thymocytes was demonstrated by hybridization of the KGF probe with
cytospot preparations of unfractionated thymocytes, where the
heterogeneity of labeling was evident (inset, Figure 2B).
To assess KGF expression by thymocytes at the protein level, serum-free medium conditioned by thymocytes was subjected to immunoblot analysis with polyclonal anti-human KGF antibodies. As shown in Figure 2C, thymocyte-conditioned medium contained proteins of ~27, ~20, and ~16 kDa detected with this antibody. The 20-kDa species has an apparent molecular weight similar to that of bacterially expressed recombinant human KGF, and the largest band had a mobility similar to that reported for KGF expressed by mammalian cells. KGF fragments with apparent molecular weights in the 16-18 kDa range have been observed before.10,32 Interestingly, processed forms in this size range are several-fold more active than the unprocessed form in keratinocyte mitogenic assays (J.S.R., unpublished observations, 1995, and Ron et al33). The lack of reactivity displayed by unconditioned medium demonstrated the thymocyte origin of the immunoreactive material. The anti-KGF antibody failed to react with recombinant FGF-1 or FGF-2, but did exhibit some cross-reactivity with recombinant FGF-10, which had an electrophoretic mobility distinct from KGF. Probing a replicate blot with polyclonal anti-FGF-1 antibodies detected recombinant FGF-1, but showed no reactivity with thymocyte-conditioned medium (data not shown). In contrast to KGF, intrathymic expression of FGF-10, a closely related FGF family member, was not detectable by RT-PCR analysis of either unfractionated neonatal, 1-week postnatal thymocytes (Figure 2D) or sorted adult thymocyte populations (no signal at 50 cycles in real-time PCR assay; data not shown). RT-PCR analysis of whole thymus consistently generated a signal in the fetal and postnatal thymus, indicating that the FGF-10 expression exhibited in whole thymus is a contribution of stromal cells, presumably mesodermally derived, or hematogenous cells that are not liberated by mechanical dissociation of the thymus. We also noted an FGF-10 signal in unfractionated fetal thymocytes that declined with increasing gestational age. The source of this signal, either fetal thymocytes or contaminating stromal cells, remains to be determined. Exogenous KGF perturbs several functional characteristics of thymic epithelium in vitro To assess the direct effects of KGF on thymic epithelium without the confounding activity of thymocytes, fetal thymic lobes were cultured in medium containing deoxyguanosine to deplete hematogenous elements, and then cultured an additional 48 hours with or without recombinant KGF prior to analysis. As a consequence of DOG treatment and depletion of hematogenous elements, the cultured lobes consisted of compact epithelium, as reflected by E-cadherin expression (Figure 3Ai-ii) that displayed prominent epithelial cysts irrespective of the presence of KGF (asterisks). Such structures have been described in the normal thymus and are more prominent in mice that display an early arrest in thymocyte development.34 The majority of the TE in both the KGF-treated and control thymic lobes displayed a cortical phenotype (Figure 3Aiii,iv). The medullary TE compartment in the control lobes was organized predominately in discrete foci, with surrounding scattered cells. In KGF-treated lobes, these discrete foci of medullary TE failed to develop, and the medullar compartment was more diffusely organized (compare panels iii and iv of Figure 3A). Additional comparison of the organization of the medullary compartment in DOG-treated lobes and intact thymic lobes revealed that KGF exposure partially restored the medullary organization to more closely resemble that of intact lobes cultured in medium where thymocyte-stromal cell cross-talk could occur (compare Figure 3Aiii and 3Aiv with Figure 5C).
We also assessed the impact of exogenous KGF on the production of cytokines by the DOG-treated thymic lobes by RT-PCR analysis. As shown in Figure 3B, treatment with KGF resulted in elevated expression of interleukin 6 (IL-6) and thymic stromal-derived lymphopoietin (TSLP),35 while having no effect on levels of IL-7 or stem cell factor (data not shown). Exogenous KGF perturbs thymocyte development and cortical TE function in vitro To evaluate the effect of KGF on the ability of the thymus to support thymocyte development in vitro, intact fetal thymic lobes were cultured in the presence or absence of exogenous KGF. In the presence of exogenous KGF, the lobes displayed slightly reduced cellularity (~80% of control lobes cultured in medium alone) accompanied by a consistent decrease in the representation of CD4 SP thymocyte in these cultures, ranging from 50% to 75%. The CD8 SP population was variably affected (Figure 4A), but due to the meager representation of mature CD8 SP thymocytes in FTOC, the impact of KGF on their development could not be reliably assessed. Other members of the FGF family (FGF1, FGF2, FGF4, or FGF10) did not have this activity (data not shown). This effect of KGF was dose-dependent; with modest effects evident at concentrations of 10 ng/mL (data not shown). Prolonged culture in the presence of KGF (15 days) resulted in more profound reductions in the CD4+8 thymocyte subset and
reduced cellularity, indicating that exogenous KGF was not simply
slowing the tempo of their development (data not shown). These
alterations in the representation of CD4+8
thymocytes were not accompanied by detectable alteration of CD69 or CD5
expression (data not shown). Such a phenotype is consistent with a
decreased efficiency of positive selection of CD4 thymocytes.
Efficient positive selection of CD4 thymocytes is dependent on
presentation of class II MHC-peptide complexes by cortical TE and is
impaired when thymic cortical epithelial cells lack Ii or Cat
L.36-39 Cat L previously has been shown to play a critical role in generation of peptide-MHC class II complexes in thymic cortical epithelial cells.39 There was a modest reduction
in the intensity of anti-MHC class II labeling in the KGF-treated lobes (compare Figure 3Avii,viii). The in situ demonstration of reduced
MHC class II expression by thymic stromal cells was confirmed with flow
cytometric analysis of enzymatically dissociated thymic lobes. Figure
4B shows that exposure to KGF reduced MHC class II expression, where
few cells expressing high levels of MHC class II+ were
detected. (For comparison, the expression of MHC class II expression by
a thymic stromal cell line in the presence or absence of prior
interferon Exogenous KGF expands the medullary compartment in vitro and in
vivo and restores medullary-type chemokine message levels in
RAG2 /
thymocytes, we hypothesized that reduced FGFR2IIIb signaling might be a
factor contributing to the hypoplastic medullary compartment in the
RAG2/ thymus. As shown here by the expression pattern
of 3G10 staining as a marker of medullary TE, the medullary compartment
in the RAG2/ thymus consisted of isolated cords of
cells previously demonstrated in association with vascular elements
(Figure 5Ai).40 We also assessed the expression of several chemokines considered to be selectively produced by medullary epithelium in the normal
thymus2,41,42 as another parameter of the epithelial
environment of the RAG2/ thymus and found that message
levels for macrophage-derived chemokine (MDC), EB-11 ligand (ELC),
secondary lymphoid tissue chemokine (SLC), and eotaxin were
dramatically reduced in the RAG2/ thymus, while message
levels for thymus-derived chemokine (TECK) and stromal derived factor-1
(SDF-1), 2 chemokines without preferential medullary
expression,43,44 were comparable in wild-type and RAG2/ thymus samples (Figure 5Aii).
To test the hypothesis that low intrathymic levels of KGF contribute to
the medullary phenotype in RAG2 This KGF-mediated expansion of the medullary epithelial compartment in
RAG2
Evidence that FGFR2IIIb signaling influences thymic development has come largely from disruption of this signaling pathway by targeted deletion of this receptor isoform or one of the primary ligands, FGF10.14 Here we analyze the reciprocal situation, where stimulation of thymic FGFR2IIIb signaling by administration of recombinant KGF affects both the developmental and functional activity of TE. Given unremarkable thymic phenotype of KGF-deficient mice,47 the response of thymic epithelium to exogenous KGF described here indicates that KGF probably plays a redundant role in thymic development/function and that the level of FGFR2IIIb signaling may be more important than the levels of an individual ligand for these receptors. The basis for the nonequivalent actions of KGF and FGF-10 observed in organ cultures is unknown. Since KGF and FGF10 differ in their dependence on low-affinity interactions with proteoglycans to signal via FGFR2IIIb,48 the differential effects of KGF and FGF10 reported here may reflect differences in presentation of these mediators by extracellular matrix components within the thymus. One consequence of elevated KGF signaling through the FGFR2IIIb
receptor in thymic epithelial cells is the reduced expression of class
II MHC, invariant chain, and cathepsin L. Given the dramatic reduction
of class II-MHC-mediated positive selection observed in mice bearing
targeted deletions of these molecules,36,39,49 this effect
of KGF on TE presents a potential explanation for the impaired
generation of CD4+8 The medullary TE compartment is also a target of FGFR2IIIb signaling
activity. The KGF-mediated expansion of the medullary compartment in
intact FTOC but not DOG-treated lobes may indicate a requirement for
additional thymocyte-derived signals for expansion of the medullary TE
compartment, or may reflect the kinetics of medullary expansion, since
the intact lobes were cultured for longer periods of time. In the
RAG2 Chemokine expression, as a functional parameter of TE, was also found
to be a target of KGF activity. In the hypoplastic medullary TE
compartment in the RAG2 The finding here of expression of KGF by While thymocytes contribute to intrathymic levels of KGF, they are not the sole source of KGF within the thymus. Although not demonstrated here, fibroblasts also are a source of KGF in the thymus as they are in other tissues (Revest et al14 and vida infra). The well-established requirement for fibroblasts in thymic organogenesis and thymic function in vitro5,54 likely reflects their elaboration of mediators of epithelial/mesenchymal interaction, including KGF. Thymocyte production of KGF may provide a feedback mechanism to more precisely regulate levels of this factor within the intrathymic milieu. Considering thymocytes as a mesenchymal derivative, their ability to
affect TE growth and/or differentiation through the elaboration of KGF
has ample precedent in the literature. In a number of tissues, including accessory male reproductive organs,55
lung,56 and epidermal epithelium,31,57 KGF
elaborated by mesenchymal cells in connective tissue has been
implicated in the regulation of epithelial cell growth and/or
differentiation (reviewed in Mason et al11 and Rubin et
al58). There are probably redundant mechanisms to control
these processes in vivo, and the relatively mild thymic and systemic
phenotype exhibited by KGF The findings here, that exogenous KGF can affect several parameters of the thymic environment, are relevant to studies investigating the use of KGF to enhance alloengraftment efficiency in bone marrow transplantation models. In addition to blunting epithelial damage from conditioning regimens,60,61 exogenous KGF also appears to enhance allogeneic engraftment through other mechanisms.62 The data presented here suggest that KGF may do so by altering the functional activity of the thymic epithelium. From a clinical standpoint, these findings also raise a cautionary note by demonstrating that exogenous KGF may affect T-cell receptor repertoire selection.
We thank Dr Bevan for comments on the manuscript, Amgen for generously providing recombinant human KGF, and Ms Dawna Dennis for expert technical assistance.
Submitted April 3, 2002; accepted June 11, 2002.
Prepublished online as Blood First Edition Paper, June 21, 2002; DOI 10.1182/blood-2002-04-1036.
Supported by National Institutes of Health (grant numbers AI24137, AG04360, and AI24206) and the Howard Hughes Medical Institute.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: Andrew G. Farr, University of Washington, School of Medicine, Department of Biological Structure, Box 357420, Seattle, WA 98195-7420; e-mail: farr{at}u.washington.edu.
1.
Hoffmann MW, Allison J, Miller JF.
Tolerance induction by thymic medullary epithelium.
Proc Natl Acad Sci U S A.
1992;89:2526-2530
2.
Campbell JJ, Pan J, Butcher EC.
Cutting edge: developmental switches in chemokine responses during T cell maturation.
J Immunol.
1999;163:2353-2357 3. van Ewijk W, Shores EW, Singer A. Crosstalk in the mouse thymus. Immunol Today. 1994;15:214-217[CrossRef][Medline] [Order article via Infotrieve]. 4. Auerbach R. Morphogenetic interactions in the development of the mouse thymus gland. Developmental Biology. 1960;2:271-284[CrossRef][Medline] [Order article via Infotrieve]. 5. Anderson G, Jenkinson EJ, Moore NC, Owen JJ. MHC class II-positive epithelium and mesenchyme cells are both required for T-cell development in the thymus. Nature. 1993;362:70-73[CrossRef][Medline] [Order article via Infotrieve].
6.
Suniara RK, Jenkinson EJ, Owen JJ.
An essential role for thymic mesenchyme in early T cell development.
J Exp Med.
2000;191:1051-1056 7. Shinohara T, Honjo T. Epidermal growth factor can replace thymic mesenchyme in induction of embryonic thymus morphogenesis in vitro. Eur J Immunol. 1996;26:747-752[Medline] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
-lineage thymocytes contribute to
intrathymic levels of KGF. Thymocyte expression of KGF is
developmentally regulated, being undetectable in
CD3
thymocytes in FTOC cultured with
exogenous KGF. We speculate that the reduced levels of surface class II
expression in response to elevated levels of KGF may reflect limiting
amounts of invariant chain, based on the important role of invariant
chain in the regulation of class II MHC transport.
-lineage T
cells