Intestinal T Cells Develop in Mice Lacking Thymus, All Lymph Nodes, Peyer’s Patches, and Isolated Lymphoid Follicles
http://www.100md.com
免疫学杂志 2005年第4期
Abstract
Through analysis of athymic (nu/nu) mice carrying a transgenic gene encoding GFP instead of RAG-2 product, it has recently been reported that, in the absence of thymopoiesis, mesenteric lymph nodes and Peyer’s patches (PP) but not gut cryptopatches are pivotal birthplace of mature T cells such as the thymus-independent intestinal intraepithelial T cells (IEL). To explore and evaluate this important issue, we generated nu/nu mice lacking all lymph nodes (LN) and PP by administration of lymphotoxin- receptor-Ig and TNF receptor 55-Ig fusion proteins into the timed pregnant nu/+ mice that had been mated with male nu/nu mice (nu/nu LNP– mice). We also generated nu/nu aly/aly (aly, alymphoplasia) double-mutant mice that inherently lacked all LN, PP, and isolated lymphoid follicles. Although -IEL were slightly smaller in number than those in nu/nu mice, substantial colonization of -IEL was found to take place in the intestinal epithelia of nu/nu LNP– and nu/nu aly/aly mice. Notably, the population size of a major CD8+ -IEL subset was maintained, the use of TCR--chain variable gene segments by these -IEL was unaltered, and the development of cryptopatches remained intact in these nu/nu LNP– and nu/nu aly/aly mice. These findings indicate that all LN, including mesenteric LN, PP, and isolated lymphoid follicles, are not an absolute requirement for the development of -IEL in athymic nu/nu mice.
Introduction
Over the past 2 decades, it has been revealed that numerous intestinal intraepithelial T cells (IEL)3 have cellular and behavioral characteristics distinct from those of thymus-derived peripheral T cells (1, 2, 3, 4, 5, 6, 7, 8). In mice, IEL are enriched with TCR- T cells (-IEL) (9, 10), and virtually all -IEL and many -IEL, unlike thymus-derived CD8 T cells that use the -chain as part of their CD3 complex, express the unique CD8 homodimer (11, 12, 13, 14) and can use the FcR-chain in place of the -chain (15, 16, 17). Along these findings, growing evidence has indicated thymus-independent (TI) development of such CD8-expressing IEL (TI-IEL) (5, 7, 11, 12, 18). Detection of RAG-1 and RAG-2 transcripts (12, 19, 20, 21, 22) and identification of a small number of T-lineage-committed TCR– lymphocytes in IEL from wild-type mice (2, 12, 19, 20, 23, 24, 25) supported the concept of localized development of IEL in the epithelial layer in situ. However, it should be pointed out that the original view of extrathymic generation of CD8+ -IEL is now inconsistent with the results of recent studies in which the thymus-dependent generation of every -IEL, including the CD8-expressing subset, is unequivocally demonstrated (26, 27).
Our search (28) for anatomical sites of IEL generation revealed multiple tiny clusters filled with 1000 c-Kit+IL-7R+Lin– (Lin, lineage markers) lymphohemopoietic cells in the lamina propria (LP) of the intestinal crypt (cryptopatches (CP)). Data obtained through a series of CP studies strongly indicated that CP were essential sites for the extrathymic development of precursor T cells destined to become TI-IEL (22, 28, 29, 30). Specifically, the presence of both TCR- and - germline transcripts in the c-Kit+IL-7R+Lin– CP lymphocytes (30) has emphasized that various DNA recombination enzymes are able to approach these chromosomal segments to commence the region-specific recombinations (31, 32, 33). On the whole, these findings lend strong support to the idea that T lineage-committed precursors, which match the developmental stage of triple-negative c-KithighCD44+CD25low/– thymocytes before pre-T gene transcription (34, 35), but after expression of CD3-specific mRNA (35, 36), are present in gut CP (30). One impediment to this conclusion has been the detection of a marginal level of RAG-2 transcripts for CP lymphocytes (22). However, the analysis of athymic (nu/nu), bone marrow (BM) chimeric mice revealed that the development of donor BM-derived TI-IEL proceeded through several consecutive steps (30). BM-derived TCR– IEL first appeared within villous epithelia overlying the regenerated CP filled with BM-derived c-Kit+IL-7R+Lin– cells. These TCR– IEL subsequently emerged throughout the epithelia, and thereafter, conversion of TCR– to TCR+ IEL, the final step, took place very slowly. These results in conjunction with above-mentioned findings (2, 12, 19, 20, 21, 22, 23, 24, 25) have led us to conclude that TI-IEL complete their late maturational events, such as RAG-mediated TCR gene rearrangement, at a very slow rate in the epithelial layer in situ.
Recently, however, a new scenario for the extrathymic development of TI-IEL in nu/nu mice was described (37). By assessing RAG-2 expression in transgenic (Tg) nu/nu mice carrying a bacterial artificial chromosome encoding a GFP reporter instead of RAG-2, it was demonstrated that extrathymic T lymphopoiesis occurred mainly in mesenteric lymph nodes (MLN) and less in Peyer’s patches (PP), but not in CP (37). To evaluate these new and important findings, we generated nu/nu mice that lacked all LN and PP by administration of lymphotoxin- receptor (LT-R)-Ig and TNF-R55-Ig fusion proteins into pregnant nu/+ mice (38) and double-mutant nu/nu aly/aly (aly, alymphoplasia) mice that lacked all LN, PP, as well as newly identified intestinal isolated lymphoid follicles (ILF) (39). We confirmed that these two kinds of mice harbored numerous -IEL in the epithelial compartments of the small intestines and were found to retain gut CP. The significance of these findings is discussed from the viewpoint that all LN, PP, and ILF are dispensable anatomical sites for the generation of TI-IEL in the athymic nu/nu condition.
Materials and Methods
Mice
BALB/cA Jcl nu/nu (nu/nu), BALB/cA Jcl nu/+ (nu/+), aly/aly Jcl mutant and C.B-17/Icr Jcl scid/scid (scid/scid) mice were purchased from CLEA Japan (Tokyo, Japan). TCR- (KN6)-Tg mice on the BALB/c background were described previously (40). Female KN6-Tg mice were crossed with scid/scid mice to generate KN6 scid/+ mice, then they were backcrossed with scid/scid mice to obtain KN6 scid/scid mice. The presence of KN6-Tg was determined by PCR analysis of tail DNA with a set of primers to the KN6 Tg (5'-CAGATCCTTCCAGTTCATCC-3' and 5'-CAGTCACTTGGGTTCCTTGTCC-3'), and the homozygous scid/scid genotype was determined by the absence of TCR-+ T cells in PBL. We generated transplacentally manipulated nu/nu and nu/+ mice that lack all LN and PP according to essentially the same method described previously (38). In brief, timed-pregnant nu/+ mice that had been mated with male nu/nu mice were i.v. injected with 200 μg of both LT-R-Ig and TNFR-55-Ig fusion proteins on gestational days 13 and 16. All mice used for experiments were between 8 and 18 wk of age, and absence of a thymus in various athymic mice was checked at necropsy. All animal procedures described in this study were performed in accordance with the guidelines for animal experiments of Keio University School of Medicine.
Production of nu/nu aly/aly mice and genotyping of aly mutation
Because nu/nu and aly/aly mothers are incapable of nursing the neonates, we used an in vitro fertilization technique (41) to produce (nu/nuxaly/aly)F1 hybrid mice, then these heterozygous nu/+ aly/+ mice were intercrossed to obtain nu/nu aly/+, nu/nu aly/aly, and nu/+ aly/+ nu/+ aly/aly littermates. To determine aly/aly, aly/+, and +/+ alleles, the TaqMan assay of tail DNA was performed using the ABI PRISM 7000 sequence detection system (PerkinElmer) as previously described (42). The primer sequences for aly were 5'-GCCTACTGACATCCCGAGCTA-3' (forward primer) and 5'-GCAGGACTGGGCTGGAAGA-3' (reverse primer). The oligonucleotide probe corresponding mutant aly allele was 5'-AGACCGTACTGTTGAAG-3' (FAM labeled), and the oligonucleotide probe corresponding wild-type allele was 5'-AGACCGTACCGTTGAA-3' (VIC labeled). Underlining in sequences indicates point mutation. The 3' end of each probe carried the quencher that suppressed the fluorescence of the reporter dyes. Each DNA sample was amplified with the TaqMan Universal master mixture containing AmpliTaq Gold DNA polymerase according to the manufacturer’s instructions (Applied Biosystems). PCR conditions were 2 min at 50°C, 10 min at 95°C, 15 s at 95°C, and 1 min at 60°C for 40 cycles. During PCR, fluorescence developed when the oligonucleotide hybridized to perfectly matching DNA, and the exonuclease activity of Taq polymerase separated the quencher from the reporter dye. After PCR, the fluorescence yield for the two different dyes was measured and presented in a two-dimensional graph.
Antibodies
The following mAbs, described previously (22, 28, 29, 30, 39), were used. For immunohistochemical and immunofluorescence stainings: anti- (GL-3), anti-c-Kit (ACK-2), anti-B220 (RA3-6B2), and anti-IgA (C10-3) were used. For flow cytometric analysis, FITC-conjugated anti- (H57-597), anti- (GL-3), anti-V1 (2.11; gift from Dr. S. Tonegawa, Center for Learning and Memory, MIT, Cambridge, MA), anti-V4 (UC3-10A6), anti-V7 (GL-1; gift from Dr. L. Lefrancois, Department of Medicine, Division of Immunology, University of Connecticut Health Center, Farmington, CT), anti-CD4 (GK 1.5), biotinylated anti- (GL-3), anti-B220 (RA3-6B2), anti-CD8 (53-6.7), and anti-c-kit (ACK-2), and PE-conjugated anti-CD8 (53-5.8) and anti-CD4 (GK 1.5) were used.
Immunohistochemical procedure
Longitudinally opened small intestine, 10 mm in length, was pasted on a filter paper to form a horizontal section and then embedded in OCT compound (Tissue-Tek; Miles) at –80°C. The tissue segments were sectioned with a cryostat at 6 μm, and sections were preincubated with Block-Ace (Dainippon Pharmaceutical) to block nonspecific binding of mAbs. The sections were then incubated with hamster (anti-) or rat (anti-c-Kit, anti-B220, or anti-IgA) mAb for 30 min at 37°C and rinsed three times with PBS, followed by incubation with biotin-conjugated goat anti-hamster IgG Ab (5 μg/ml; Cedarlane Laboratories) or with biotin-conjugated goat anti-rat IgG (5 μg/ml; Cedarlane Laboratories). Subsequently, the sections were washed three times with PBS, then incubated with avidin-biotin peroxidase complexes (Vectastain ABC kit; Vector Laboratories). Histochemical color development was achieved with Vectastain 3,3'-diaminobenzidine substrate kit (Vector Laboratories) according to the manufacturer’s instructions. Finally, the sections were counterstained with hematoxylin for microscopy. Endogenous peroxidase activity was blocked with 0.3% H2O2 and 0.1% NaN3 in distilled water for 10 min at room temperature. Tissue sections incubated with either nonimmune hamster serum or isotype-matched normal rat IgG showed only minimal background staining.
Immunofluorescence procedure
Tissue segments from thymus, spleen, inguinal LN, MLN, and PP from KN6 scid/scid mice were embedded in OCT compound at –80°C. The small intestine of KN6 scid/scid mice was longitudinally opened along the mesenteric wall, then intestine, 10 mm in length, that had been rolled to form a vertical section was embedded in OCT compound at –80°C. Cryostat tissue sections, 6-μm thick, were fixed in acetone for 10 min at room temperature, washed three times with PBS, then pretreated with Block-Ace. Subsequently, the sections were incubated with anti-c-Kit mAb (ACK-2) for 60 min at 4°C, followed by incubation with PE-conjugated goat F(ab')2 anti-rat IgG (H+L) (Invitrogen Life Technologies). The sections were then incubated with anti- mAb (GL-3) and counterstained with FITC-conjugated goat anti-hamster IgG (H+L) (Jackson ImmunoResearch Laboratories). Finally, the sections were examined under a fluorescence microscope (Axiovert 100; Carl Zeiss) equipped with an image analysis system (Signal Analytics).
Flow cytometry
IEL were isolated according to methods described previously (22). Lymphoid cells were incubated first with biotinylated mAb, then with streptavidin-PE (BD Biosciences) and FITC-conjugated second mAb. Stained cells were suspended in staining medium (Hanks’ solution without phenol red, 0.02% NaN3, and 2% heat-inactivated FBS) containing 0.5 μg/ml propidium iodide and analyzed using FACScan with CellQuest software (BD Biosciences). Dead cells were excluded by propidium iodide gating. Three-color analysis of IEL was also performed. IEL were incubated first with anti-CD8 mAb (biotinylated), then with streptavidin-Tri-Color (Caltag Laboratories). After washing, IEL were counterstained with two combinations of two PE-conjugated mAbs (anti-CD8 and anti-CD4 mAbs) and one FITC-conjugated mAb (anti-), respectively. Lymphoid cells were incubated with anti-FcR II/III mAb (2.4G2) before staining to block nonspecific binding of labeled mAbs to FcR.
Results
Development of -IEL in nu/nu mice is independent of all LN and PP
To explore whether MLN are essential anatomical sites for the generation of TI-IEL in athymic nu/nu mice (37), we generated nu/nu mice that lacked all LN and PP. Pregnant nu/+ female mice that had been mated with nu/nu male mice were injected with LT-R-Ig and TNF-R55-Ig fusion proteins according to the protocol described by Rennert et al. (38), and the presence or the absence of LN and PP was determined in the progeny at 8 wk of age under a stereomicroscope. Although PP were absent from all treated mice, markedly attenuated remnants of MLN were present in about one-fifth of them. However, in every MLN-deficient nu/nu and nu/+ offspring, the development of mandibular, axillary, inguinal, and popliteal (data not shown) LN (i.e., peripheral LN) and cervical (data not shown), iliac, and sacral LN (i.e., mucosal LN) was also ablated (LNP– mice; Fig. 1).
FIGURE 1. Development of PP and all LN in athymic nu/nu mice is ablated by in utero treatment with both LT-R-Ig and TNF-R55-Ig fusion proteins. PP (A), MLN (B), mandibular LN (C), axillary LN (D), iliac LN (E), sacral LN (F), and inguinal LN (G) are present in untreated nu/nu mice (arrowheads), but are undetectable in nu/nu mice that were treated in utero with both LT-R-Ig and TNF-R55-Ig fusion proteins (nu/nu LNP– mice). Bar, 2 mm.
Flow cytometric analysis of IEL isolated from nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice was performed using anti-TCR- and anti-TCR- mAbs. Consistent with well-established findings (3, 4, 8, 43), the proportion of -IEL to -IEL was sharply reduced, and the composition of IEL not expressing either type of TCR was expanded in athymic nu/nu conditions regardless of the presence or the absence of all LN and PP (Fig. 2A). In contrast, no significant differences in absolute numbers of IEL were observed between LNP– and control LNP+ mice (data not shown). Notably, although the population size was slightly smaller by a factor of 1.5 compared with that in IEL from control nu/nu mice, a large number of -IEL was detected in IEL from nu/nu LNP– mice (Fig. 2A).
Compartmentalization of -IEL within the epithelial layer of small intestine in nu/nu LNP– mice was also verified by immunohistochemistry (Fig. 2B). These results indicate that the development of -IEL per se is independent of thymus, all LN, and PP.
FIGURE 2. Flow cytometric analysis of IEL and immunohistochemical examination of small intestines from nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice. A, Although the population size of -IEL in nu/nu LNP– mice that lack all LN and PP is smaller than that of nu/nu mice, a substantial number of -IEL are present in the epithelial compartment of nu/nu LNP– mice, indicating that -IEL are capable of developing in the absence of thymus, all LN including MLN, and PP. Note that the composition of -IEL is reduced drastically in the athymic nu/nu condition compared with that in the euthymic nu/+ condition. B, Representative immunohistochemical visualization of -IEL in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). Although numbers of -IEL in nu/nu LNP– mice are decreased compared with those in nu/nu mice, colonization of -IEL takes place in the absence of thymus, all LN including MLN, and PP (arrowheads). C, Representative immunohistochemical verification of CP in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). An average number and an approximate mass of CP filled with c-Kit+B200+ lymphocytes in these four different mice remain almost the same. D, Representative immunohistochemical verification of ILF in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). Note that a cluster of B220+ B cells that reside in the central region of ILF is surrounded by the layer of cells expressing c-Kit molecules.
With these findings in mind, we examined whether CP and ILF were present in these in utero manipulated LNP– mice, because, in contrast to PP that are already microscopically well developed just before birth (44), organogenesis of CP (28) and ILF (39) commences in early postnatal life. In fact, it was corroborated that the development of CP filled with closely packed c-Kit+ lymphocytes (Fig. 2C) and ILF containing B220+ B cell aggregation (Fig. 2D) remained intact in these nu/nu LNP– and nu/+ LNP– mice.
Development of -IEL in nu/nu aly/aly double-mutant mice
To ascertain the universality of the above findings, we explored the development of IEL in a mutant mouse that inherently lacked thymus, all LN, and PP, because fusion protein-treated LNP– mice might possess a minute and stereomicroscopically invisible MLN, even though this possibility appeared to be remote (38). With this purpose in mind, we generated nu/nu aly/aly double-mutant mice. In accordance with the earliest description (45), nu/nu aly/aly mice were devoid of all LN and PP (data not shown) as well as thymus. Importantly, substantial colonization of -IEL in the small intestine of nu/nu aly/aly mice was verified by flow cytometric (Fig. 3A) and immunohistochemical (Fig. 3B) analyses. Furthermore, as inferred from our previous observations (28, 39), histogenesis of CP was detected (Fig. 3C), whereas development of ILF was completely blocked (Fig. 3D), in these double-mutant animals. Taking all of these results together (Figs. 2 and 3), neither thymus, all LN including MLN, PP, nor ILF is an absolute requirement for the development of -IEL.
FIGURE 3. Flow cytometric analysis of IEL and immunohistochemical examination of small intestines from nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice. A, Although the population size of -IEL in nu/nu aly/aly mice that lack all LN, PP, and ILF is smaller than that in nu/nu aly/+ mice, a significant number of -IEL are present in the epithelial compartment of nu/nu aly/aly mice, indicating that -IEL are capable of developing in the absence of thymus, all LN including MLN, PP, and ILF. Note that the composition of -IEL is reduced drastically in the athymic nu/nu condition compared with that in the euthymic nu/+ condition. B, Representative immunohistochemical verification of -IEL in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Although numbers of -IEL in nu/nu aly/aly mice are lower by a factor of 1.5–2 compared with those in nu/nu aly/+ mice, a significant colonization of -IEL takes place in the absence of thymus, all LN including MLN, PP, and ILF (arrowheads). C, Representative immunohistochemical verification of CP in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Although an average number of CP filled with c-Kit+B220– lymphocytes in these four different mice remain almost the same, an approximate mass of CP present in nu/nu aly/aly and nu/+ aly/aly mice is slightly reduced compared with that of CP present in nu/nu aly/+ and nu/+ aly/+ mice. D, Representative immunohistochemical verification of ILF in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Note that ILF are undetectable throughout the small intestinal mucosa of nu/nu aly/aly and nu/+ aly/aly mice.
In this context, it is important to determine T and B cells and IgA+ B cells that sojourn, respectively, in the spleen and LP of nu/nu LNP– and nu/nu aly/aly mice, because spleen is most likely the sole organized peripheral lymphoid tissue remaining in these animals, and nu/nu aly/aly mice lack intestinal IgA+ B cell-producing plants such as PP and ILF. In contrast to abundant B220+ B cells, mature T cells were virtually absent in the spleens of nu/nu LNP– (Fig. 4A) and nu/nu aly/aly (Fig. 4C) mice. In the small intestines, however, nu/nu LNP– mice possessed IgA+ B cells (Fig. 4B), -IEL, CP, and well-developed ILF (Fig. 2), whereas nu/nu aly/aly mice possessed -IEL (Fig. 3, A and B) and CP (Fig. 3C), but lacked IgA+ B cells (Fig. 4C) and ILF (Fig. 4D). These results indicate that all LN, including MLN, PP, and ILF, are not an absolute requirement for the generation of -IEL in nu/nu mice and that the aly mutation interferes the formation of PP and ILF, resulting in the impaired development of IgA+ B cells in villous LP.
FIGURE 4. Flow cytometric analysis of splenic T and B cells and immunohistochemical verification of intestinal IgA+ B cells in nu/nu, nu/nu LNP–, nu/nu aly/+, and nu/nu aly/aly mice. A, Although mature CD3+ T cells are nearly absent from the spleens of nu/nu and nu/nu LNP– mice, abundant B220+ B cells are detected in the spleens of these athymic animals. Colonization of IgA+ B cells in the villous LP of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice is comparable. B, Only a marginal number of mature CD3+ T cells are detected in the spleens of nu/nu aly/+ and nu/nu aly/aly mice, and a large fraction of the remaining lymphoid cells consists of B220+ B cells. In contrast, villous LP of nu/nu aly/aly and nu/+ aly/aly mice has no IgA+ B cells.
Phenotypic and V gene usage analyses of -IEL in nu/nu LNP– and nu/nu aly/aly mice
The data reported to date indicate that -IEL are potentially capable of developing in nu/nu mice lacking all LN, PP, and ILF. In this regard, however, it is reasonable to consider the possibility that -IEL generated under such harsh conditions might differ from those generated in nu/nu mice possessing all LN, PP, and ILF. To address this issue, we conducted flow cytometric analysis of -IEL isolated from nu/nu, nu/nu LNP–, and nu/nu aly/aly mice. Although absolute numbers of -IEL were lower by a factor of 2 compared with those in nu/nu mice (Fig. 5A), the composition of the major -IEL subset expressing CD8 homodimer (Fig. 5A) and the V1, V4, and V7 gene segment used by such -IEL (Fig. 5B) remained the same in both nu/nu LNP– and nu/nu aly/aly mice. Collectively, these results indicate that the absence of all LN, PP, and ILF exerts only a small effect on the phenotypic configuration of -IEL in nu/nu mice.
FIGURE 5. Flow cytometric analysis of -IEL from nu/nu, nu/nu LNP–, nu/nu aly/+, and nu/nu aly/aly mice and V gene segments used by -IEL in these athymic nu/nu mice. A, Three-color analysis was performed. Absolute numbers of double-negative (CD4–8–), single-positive (CD4+, CD8+, or CD8+), and double-positive (CD4+8+) subsets in the -IEL population were calculated on the basis of total numbers of -IEL. Data are the mean values from five mice per group. B, Two-color analysis was performed. IEL isolated from these nu/nu mice were incubated first with anti-TCR- mAb (biotinylated), then with streptavidin-PE and FITC-conjugated anti-V1, anti-V4, or anti-V7 mAb. The results are the mean ± SD of data obtained from four mice per group.
Development of c-Kit+ -IEL in scid/scid mice expressing Tg TCR-
We have shown that IL-7 produced by intestinal epithelial cells (IEC) is important for intraintestinal development of -IEL and is crucial for organization of intestinal mucosal lymphoid tissues, such as PP and CP (18). It has also been shown that c-Kit and stem cell factor (SCF) are expressed by -IEL and IEC, respectively, and signaling through c-Kit/SCF is indispensable for normal development of -IEL (46, 47). Thus, these previous (46, 47) and the present findings in conjunction with results obtained with BM chimeric mice (30), reinforce the idea that the development of -IEL takes place in the intestinal mucosa in situ. In an attempt to confirm and visualize directly the cellular events that proceed toward gut-oriented -IEL generation, i.e., c-Kit+ CP cellsc-Kit+ TCR- T cellsc-Kit– -IEL, we generated scid/scid mice expressing KN6-Tg TCR- (48). Double-immunofluorescence analysis of small intestinal tissues containing CP highlighted these presumptive cellular events. Thus, a representative picture of jejunal tissue sections from KN6 scid/scid mice (Fig. 6A) shows that a cluster of c-Kit+ cells in a CP (red) does not express TCR-, and that large numbers of -IELs (green) are present in the epithelial layer, especially in the epithelium adjacent to CP. Notably, quite a large number of lymphocytes expressing both c-Kit and TCR- molecules (yellow or orange) is also present in the epithelial and LP compartments of villi (Fig. 6A). Neither clustering c-Kit+ cells nor lymphocytes stained yellow/orange were detected in other lymphoid tissues, such as thymus, LN, PP, and spleen (data not shown). Flow cytometric analysis of cell surface c-Kit molecules on the gated T cells confirmed that a large fraction of -IEL was c-Kit positive, namely, double-positive TCR-+c-Kit+ cells (Fig. 6B, upper panel). In contrast, abundant Tg T cells residing in MLN, spleen, and thymus did not include such c-Kit-expressing, double-positive cells (Fig. 6B, lower three panels). Overall, these results support the above-described basic premise that the development of -IEL takes place in the intestinal mucosa in situ. However, because KN6-Tg TCR--expressing cells have not been detected in the CP of KN6 scid/scid mice (Fig. 6A), whether CP are essential and indispensable anatomical sites in generating -IEL remained highly contentious.
FIGURE 6. Double immunofluorescence analysis of small intestinal tissues containing CP and flow cytometric analysis of lymphocytes from KN6 scid/scid mice. A, Intestinal birthplace of intraepithelial T cells (magnification, x400). A cluster of lymphocytes in a CP is positively stained with anti-c-Kit mAb (PE), but not with anti- mAb (FITC), resulting in a red color. Conversely, many IEL, especially those adjacent to CP, are positively stained with anti- mAb (FITC), but not with anti-c-Kit mAb (PE), resulting in a green color. Note that a large number of lymphocytes present in the epithelial and LP compartments of villi are positively stained with anti-c-Kit mAb (PE) and anti- mAb (FITC), resulting in a yellow or orange color. Neither the clusters filled with c-Kit+ cells nor lymphocytes stained yellow or orange are detected in the other lymphoid organs, such as thymus, LN, and spleen. B, Expression of c-Kit molecules by Tg T cells that colonize in the intestinal epithelial, MLN, and splenic and thymic compartments of KN6 scid/scid mice. IEL, MLN cells, spleen cells, and thymocytes were incubated first with anti-TCR- mAb (biotinylated), then with streptavidin-PE and FITC-conjugated anti-c-Kit mAb. Profiles of c-Kit expression by the gated TCR-+ cell population are presented. The dark area in the upper panel depicts a FITC-positive cell profile of IEL that were stained simply with biotinylated anti-TCR- mAb and streptavidin-PE (negative control). It is evident that IEL contain a large number of double-positive (TCR-+ c-Kit+) cells, but that MLN, spleen, and thymus have no such double-positive cells.
Discussion
Because most of the numerous lymphocytes residing in the murine IEC compartment unexpectedly turned out to be T cells (IEL), additional revelation toward the end of the last century indicated that they display phenotypic and functional characteristics distinct from those of other T cell populations and that their development does not necessarily depend on the thymus (TI-IEL) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18). Likewise, several lines of information have illuminated the distinctive T cell facets of human fetal intestine (49, 50), and on the basis of RAG expression, there may also be TI-IEL that develop in human (51, 52) and rat (53) intestines. Although evidence for the vestigial lymphocyte-producing function of gut mucosa is substantial, as mentioned above (2, 5, 7, 11, 12, 18, 19, 20, 21, 22, 23, 24, 25, 49, 50, 51, 52, 53), recent studies have made it clear that CD8 does not serve as a marker for TI development of -IEL (26, 27). In contrast, because positive and/or negative selection of TCR- T cells in the thymus is not as evident (26, 27), thymic dependency for functional TCR- T cells is less obvious.
In an assessment of the expression of Tg-encoded GFP in place of RAG-2 protein, no evidence was obtained for a lymphopoietic process involving CP cells migrating into gut epithelium to undergo TCR gene rearrangement and maturation into - and -IEL even in the athymic nu/nu condition (37). Instead, MLN and, less efficiently, PP have been identified as the major extrathymic T cell-producing plants in athymic nu/nu mice, and the newly generated T cells migrate from MLN into thoracic duct lymph to reach the gut epithelia, indicating that MLN and PP are the pivotal sites in generating TI-IEL, mostly -IEL (37). In this context, the development of -IEL should be hampered in nu/nu mice that simultaneously lack MLN and PP. Our present findings argue against this scenario by showing that not only transplacentally manipulated nu/nu LNP– mice lacking all LN and PP, but also genetically defined nu/nu aly/aly mice lacking all LN, PP, and ILF harbor a substantial population of -IEL (Fig. 2, A and B, and Fig. 3, A and B), although these two different mouse models may share the same confounding factors. In contrast to what was observed in nu/nu mice, the extrathymic pathway of IEL generation was shown to be totally repressed in the euthymic condition using the same GFP RAG-2 Tg mouse model (37). The authors proposed that all IEL, including CD8+ IEL, in normal mice were the likely progeny of double-negative (DN) TCR-+ and -+ thymocytes and noted that extremely complex and unusual T cell characteristics of murine IEL with respect to their expression of various accessory, costimulation, activation, and adhesion markers (4, 5) might be brought about by the distinctive microenvironment of gut epithelium (37). In this context, for instance, many DN thymocytes somehow down-regulate cell surface expression of Thy-1 molecules, because a substantial fraction of IEL is Thy-1 negative, whereas most of them must up-regulate c-Kit molecules on the way to becoming IEL (46, 47) (Fig. 6). Even if all of those transfigurations (>20) are attributable to the inherent properties of gut epithelium, the biological significance as well as the molecular level of the mechanisms underlying such enigmatic cellular events remain highly contentious. It should also be pointed out that our recent findings (18) are inconsistent with this idea (37). T cells are absent in IL-7–/– mice due to the selective blockade of TCR- gene rearrangements (54). Using the intestinal fatty acid-binding protein (iFABP) promoter, we reinstated the expression of IL-7 to mature IEC of IL-7–/– mice (iFABP-IL7) (18). Although -IEL were restored in iFABP-IL7 mice as well as CP and PP, T cells remained absent from all tissues, including thymus, spleen, and skin. These results clearly indicate that -IEL generated in iFABP-IL7 mice are not of DN TCR-+ thymocyte origin and that the recombination of TCR- genes in TCR-precursor T cells takes place in situ with the assistance of IL-7 produced locally by IEC.
Our present findings provide compelling evidence for the development of -IEL within the intestine of nu/nu mice that lack the thymus, all LN, PP, and ILF. It should be pointed out, however, that the population size and absolute numbers of -IEL from nu/nu LNP– and nu/nu aly/aly mice are smaller than those from the corresponding control nu/nu mice (Figs. 2A, 3A, and 5A). These features would indicate that LN and PP, in fact, contribute to -IEL numbers in the nu/nu condition. In contrast to the results obtained in GFP RAG-2 Tg mouse model (37), our previous RT-PCR analysis of lymphocytes from normal euthymic mice showed that under conditions in which mRNA from 50 thymocytes displayed a strong signal for RAG-2 transcripts and mRNA from 6250 RAG-2–/– thymocytes failed to display any detectable signals, very low levels of RAG-2 transcripts were constantly detected in an amount of mRNA equivalent to 6250 IEL and CP cells (22). These findings suggest that a small minority of IEL and possibly CP cells also is undergoing TCR gene rearrangement, that they are able to do so with a minimum amount of RAG-2 transcripts, or both. Actually, we still do not know how many RAG-1 and -2 molecules per nucleus are required to drive the region-specific V(D)J recombinations of TCR genes. It is possible that the amount of mRNA encoding RAG-1 and -2 molecules required by thymocytes for the successful recombination of TCR genes may not be that large. It should also be pointed out that using cell fate mapping, almost all -IEL have recently been shown to be the progeny of immature CD4+CD8+ thymocytes (55). Furthermore, by using elegant and sophisticated approaches (55), it has been revealed that the retinoic acid-related orphan receptors (RORt) detected in fetal lymphoid tissue-inducer cells are also expressed in cells within gut CP, and that -IEL are not the progeny of such RORt-positive CP lymphocytes. Specifically, however, it has remained an open question whether a small, but significant, fraction of lymphocytes in CP does not retain RORt molecules or whether almost all CP lymphocytes express RORt (55). In any event, to substantiate the intraintestinal development of -IEL in these nu/nu LNP– and nu/nu aly/aly mice, clear identification of T cells undergoing TCR gene rearrangement in the gut mucosa appears to be of critical importance.
Acknowledgments
We are grateful to Dr. L. Lefrancois for his critical reading of the manuscript. We thank N. Hosaka and M. Mori for their excellent technical assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Grant-in-Aid for Creative Scientific Research (13GS0015); the Japan Society for the Promotion of Science; a Grant-in-Aid for Scientific Research on Priority Areas A; a Grant-in-Aid for the 21st Century of Excellence Program entitled Understanding and Control of Life’s Function via Systems Biology (Keio University); the Special Coordination Fund for Promoting Science and Technology, Ministry of Education, Culture, Sports, Science, and Technology; and Health Science Research Grants from the Ministry of Health, Labor, and Welfare.
2 Address correspondence and reprint requests to Dr. Hiromichi Ishikawa, Department of Microbiology and Immunology, Keio University School of Medicine, Shinjuku-ku, Tokyo 160-8582, Japan. E-mail address: h-ishika@sc.itc.keio.ac.jp
3 Abbreviations used in this paper: IEL, intestinal intraepithelial T cell; aly, alymphoplasia; BM, bone marrow; CP, cryptopatch; DN, double negative; iFABP, intestinal fatty acid-binding protein; IEC, intestinal epithelial cell; ILF, isolated lymphoid follicle; Lin, lineage marker; LNP–, lymph node- and Peyer’s patch deficient; LP, lamina propria; LT, lymphotoxin; LT-R, lymphotoxin- receptor; MLN, mesenteric lymph node; PP, Peyer’s patch; RORt, retinoic acid-related orphan receptor; SCF, stem cell factor; Tg, transgenic; TI, thymus independent.
Received for publication August 9, 2004. Accepted for publication December 10, 2004.
References
Rocha, B., D. Guy-Grand, P. Vassalli. 1995. Extrathymic T cell differentiation. Curr. Opin. Immunol. 7:235.
Mowat, A. M., J. L. Viney. 1997. The anatomical basis of intestinal immunity. Immunol. Rev. 156:145.
Klein, J. R.. 1998. Thymus-independent development of gut T cells. Chem. Immunol. 71:88.
Lefrancois, L., L. Puddington. 1999. Basic aspects of intraepithelial lymphocyte immunobiology. P. L. Ogra, and J. Mestecky, and M. E. Lamm, and W. Strober, and J. Bienenstock, and J. R. McGhee, eds. Mucosal Immunology 413. Academic Press, San Diego.
Aranda, R., B. C. Sydora, M. Kronenberg. 1999. Intraepithelial lymphocytes: function. P. L. Ogra, and J. Mestecky, and M. E. Lamm, and W. Strober, and J. Bienenstock, and J. R. McGhee, eds. Mucosal Immunology 429. Academic Press, San Diego.
Poussier, P., M. Julius. 1999. Speculation on the lineage relationships among CD4–8+ gut-derived T cells and their role(s). Semin. Immunol. 11:293.
Hayday, A., E. Theodoridis, E. Ramsburg, J. Shires. 2001. Intraepithelial lymphocytes: exploring the Third Way in immunology. Nat. Immunol. 2:997.
Guy-Grand, D., P. Vassalli. 2002. Gut intraepithelial lymphocyte development. Curr. Opin. Immunol. 14:255.
Goodman, T., L. Lefrancois. 1988. Expression of the - T-cell receptor on intestinal CD8+ intraepithelial lymphocytes. Nature 333:855.
Bonneville, M., C. A. Janeway, K. Ito, W. Haser, I. Ishida, N. Nakanishi, S. Tonegawa. 1988. Intestinal intraepithelial lymphocytes are a distinct set of T-cells. Nature 336:479.
Lefrancois, L.. 1991. Phenotype complexity of intraepithelial lymphocyte of the small intestine. J. Immunol. 147:1746.
Guy-Grand, D., N. Cerf-Bensussan, B. Malissen, M. Malassis-Seris, C. Briottet, P. Vassalli. 1991. Two gut intraepithelial CD8+ lymphocyte populations with different T cell receptors: a role for the gut epithelium in T cell differentiation. J. Exp. Med. 173:471.
Maloy, K. J., A. M. Mowat, R. Zamoyska, I. N. Crispe. 1991. Phenotypic heterogeneity of intraepithelial T lymphocytes from mouse small intestine. Immunology 72:555
Kawaguchi, M., M. Nanno, Y. Umesaki, S. Matsumoto, Y. Okada, Z. Cai, T. Shimamura, Y. Matsuoka, M. Ohwaki, H. Ishikawa. 1993. Cytolytic activity of intestinal intraepithelial lymphocytes in germ-free mice is strain dependent and determined by T cells expressing T-cell antigen receptors. Proc. Natl. Acad. Sci. USA 90:8591.
Malissen, M., A. Gillet, B. Rocha, J. Trucy, E. Vivier, C. Boyer, F. Kontgen, N. Brun, G. Mazza, E. Spanopoulou, et al 1993. T cell development in mice lacking CD3-/ gene. EMBO J. 12:4347.
Ohno, H., T. Aoe, S. Taki, D. Kitamura, Y. Ishida, K. Rajewsky, T. Saito. 1993. Development and functional impairment of T cells in mice lacking CD3 chains. EMBO J. 12:4357.
Liu, C-P., R. Ueda, J. She, J. Sancho, B. Wang, G. Weddell, J. Loring, C. Kurahara, E. C. Dudley, A. Hayday, et al 1993. Abnormal T cell development in CD3–/– mutant mice and identification of a novel T cell population in the intestine. EMBO J. 12:4863
Laky, K., L. Lefrancois, E. G. Lingenheld, H. Ishikawa, J. M. Lewis, S. Olson, K. Suzuki, R. E. Tigelaar, L. Puddington. 2000. Enterocyte expression of IL-7 induces development of T cells and Peyer’s patches. J. Exp. Med. 191:1569.
Guy-Grand, D., C. V. Broecke, C. Briottet, M. Malassis-Seris, F. Selz, P. Vassalli. 1992. Different expression of the recombination activity gene RAG-1 in various populations of thymocytes, peripheral T cells and gut thymus-independent intraepithelial lymphocytes suggests two pathways of T cell receptor rearrangement. Eur. J. Immunol. 22:505.
Lin, T., G. Matsuzaki, H. Yoshida, N. Kobayashi, H. Kenai, K. Omoto, K. Nomoto. 1994. CD3–CD8+ intestinal intraepithelial lymphocytes (IEL) and the extrathymic development of IEL. Eur. J. Immunol. 24:1080.
Boll, G., A. Rudolphi, S. Spieb, J. Reimann. 1995. Regional specialization of intraepithelial T cells in the murine small and large intestine. Scand. J. Immunol. 41:103.
Oida, T., K. Suzuki, M. Nanno, Y. Kanamori, H. Saito, E. Kubota, S. Kato, M. Itoh, S. Kaminogawa, H. Ishikawa. 2000. Role of gut cryptopatches in early extrathymic maturation of intestinal intraepithelial T cells. J. Immunol. 164:3616.
Hamad, M., M. Whetsell, J. Wang, J. R. Klein. 1997. T cell progenitors in the murine small intestine. Dev. Comp. Immunol. 21:435.
Page, S. T., L. Y. Bogatzki, J. A. Hamerman, C. H. Sweenie, P. J. Hogarth, M. Malissen, R. M. Perlmutter, A. M. Pullen. 1998. Intestinal intraepithelial lymphocytes include precursors committed to the T cell receptor lineage. Proc. Natl. Acad. Sci. USA 95:9459.
Woodward, J., E. Jenkinson. 2001. Identification and characterization of lymphoid precursors in the murine intestinal epithelium. Eur. J. Immunol. 31:3329.
Leishman, A. J., L. Gapin, M. Capone, E. Palmer, H. R. MacDonald, M. Kronenberg, H. Cheroutre. 2002. Precursors of functional MHC class I- or class II-restricted CD8+ T cells are positively selected in the thymus by agonist self-peptides. Immunity 16:355.
Cheroutre, H.. 2004. Starting at the beginning: new perspectives on the biology of mucosal T cells. Annu. Rev. Immunol. 22:217.
Kanamori, Y., K. Ishimaru, M. Nanno, K. Maki, K. Ikuta, H. Nariuchi, H. Ishikawa. 1996. Identification of novel lymphoid tissues in murine intestinal mucosa where clusters of c-kit+ IL-7R+ Thy1+ lympho-hemopoietic progenitors develop. J. Exp. Med. 184:1449.
Saito, H., Y. Kanamori, T. Takemori, H. Nariuchi, E. Kubota, H. Takahashi-Iwanaga, T. Iwanaga, H. Ishikawa. 1998. Generation of intestinal T cells from progenitors residing in gut cryptopatches. Science 280:275.
Suzuki, K., T. Oida, H. Hamada, O. Hitotsumatsu, M. Watanabe, T. Hibi, H. Yamamoto, E. Kubota, S. Kaminogawa, H. Ishikawa. 2000. Gut cryptopatches: direct evidence of extrathymic anatomical sites for intestinal T lymphopoiesis. Immunity 13:691.
Alt, F. W., T. K. Blackwell, G. D. Yancopoulos. 1987. Development of the primary antibody repertoire. Science 238:1079.
Hempel, W. M., I. Leduc, N. Mathieu, R. K. Tripathi, P. Ferrier. 1998. Accessibility control of V(D)J recombination: lessons from gene targeting. Adv. Immunol. 69:309.
Ye, S.-K., K. Maki, T. Kitamura, S. Sunaga, K. Akashi, J. Domen, I. L. Weissman, T. Honjo, K. Ikuta. 1999. Induction of germline transcription in the TCR locus by Stat5: implications for accessibility control by the IL-7 receptor. Immunity 11:213.
Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H. J. Fehling, H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science 266:1208.
Wilson, A., H. R. MacDonald. 1995. Expression of genes encoding the pre-TCR and CD3 complex during thymus development. Int. Immunol. 7:1659.[(Satoshi Nonaka, Tomoaki N)
Through analysis of athymic (nu/nu) mice carrying a transgenic gene encoding GFP instead of RAG-2 product, it has recently been reported that, in the absence of thymopoiesis, mesenteric lymph nodes and Peyer’s patches (PP) but not gut cryptopatches are pivotal birthplace of mature T cells such as the thymus-independent intestinal intraepithelial T cells (IEL). To explore and evaluate this important issue, we generated nu/nu mice lacking all lymph nodes (LN) and PP by administration of lymphotoxin- receptor-Ig and TNF receptor 55-Ig fusion proteins into the timed pregnant nu/+ mice that had been mated with male nu/nu mice (nu/nu LNP– mice). We also generated nu/nu aly/aly (aly, alymphoplasia) double-mutant mice that inherently lacked all LN, PP, and isolated lymphoid follicles. Although -IEL were slightly smaller in number than those in nu/nu mice, substantial colonization of -IEL was found to take place in the intestinal epithelia of nu/nu LNP– and nu/nu aly/aly mice. Notably, the population size of a major CD8+ -IEL subset was maintained, the use of TCR--chain variable gene segments by these -IEL was unaltered, and the development of cryptopatches remained intact in these nu/nu LNP– and nu/nu aly/aly mice. These findings indicate that all LN, including mesenteric LN, PP, and isolated lymphoid follicles, are not an absolute requirement for the development of -IEL in athymic nu/nu mice.
Introduction
Over the past 2 decades, it has been revealed that numerous intestinal intraepithelial T cells (IEL)3 have cellular and behavioral characteristics distinct from those of thymus-derived peripheral T cells (1, 2, 3, 4, 5, 6, 7, 8). In mice, IEL are enriched with TCR- T cells (-IEL) (9, 10), and virtually all -IEL and many -IEL, unlike thymus-derived CD8 T cells that use the -chain as part of their CD3 complex, express the unique CD8 homodimer (11, 12, 13, 14) and can use the FcR-chain in place of the -chain (15, 16, 17). Along these findings, growing evidence has indicated thymus-independent (TI) development of such CD8-expressing IEL (TI-IEL) (5, 7, 11, 12, 18). Detection of RAG-1 and RAG-2 transcripts (12, 19, 20, 21, 22) and identification of a small number of T-lineage-committed TCR– lymphocytes in IEL from wild-type mice (2, 12, 19, 20, 23, 24, 25) supported the concept of localized development of IEL in the epithelial layer in situ. However, it should be pointed out that the original view of extrathymic generation of CD8+ -IEL is now inconsistent with the results of recent studies in which the thymus-dependent generation of every -IEL, including the CD8-expressing subset, is unequivocally demonstrated (26, 27).
Our search (28) for anatomical sites of IEL generation revealed multiple tiny clusters filled with 1000 c-Kit+IL-7R+Lin– (Lin, lineage markers) lymphohemopoietic cells in the lamina propria (LP) of the intestinal crypt (cryptopatches (CP)). Data obtained through a series of CP studies strongly indicated that CP were essential sites for the extrathymic development of precursor T cells destined to become TI-IEL (22, 28, 29, 30). Specifically, the presence of both TCR- and - germline transcripts in the c-Kit+IL-7R+Lin– CP lymphocytes (30) has emphasized that various DNA recombination enzymes are able to approach these chromosomal segments to commence the region-specific recombinations (31, 32, 33). On the whole, these findings lend strong support to the idea that T lineage-committed precursors, which match the developmental stage of triple-negative c-KithighCD44+CD25low/– thymocytes before pre-T gene transcription (34, 35), but after expression of CD3-specific mRNA (35, 36), are present in gut CP (30). One impediment to this conclusion has been the detection of a marginal level of RAG-2 transcripts for CP lymphocytes (22). However, the analysis of athymic (nu/nu), bone marrow (BM) chimeric mice revealed that the development of donor BM-derived TI-IEL proceeded through several consecutive steps (30). BM-derived TCR– IEL first appeared within villous epithelia overlying the regenerated CP filled with BM-derived c-Kit+IL-7R+Lin– cells. These TCR– IEL subsequently emerged throughout the epithelia, and thereafter, conversion of TCR– to TCR+ IEL, the final step, took place very slowly. These results in conjunction with above-mentioned findings (2, 12, 19, 20, 21, 22, 23, 24, 25) have led us to conclude that TI-IEL complete their late maturational events, such as RAG-mediated TCR gene rearrangement, at a very slow rate in the epithelial layer in situ.
Recently, however, a new scenario for the extrathymic development of TI-IEL in nu/nu mice was described (37). By assessing RAG-2 expression in transgenic (Tg) nu/nu mice carrying a bacterial artificial chromosome encoding a GFP reporter instead of RAG-2, it was demonstrated that extrathymic T lymphopoiesis occurred mainly in mesenteric lymph nodes (MLN) and less in Peyer’s patches (PP), but not in CP (37). To evaluate these new and important findings, we generated nu/nu mice that lacked all LN and PP by administration of lymphotoxin- receptor (LT-R)-Ig and TNF-R55-Ig fusion proteins into pregnant nu/+ mice (38) and double-mutant nu/nu aly/aly (aly, alymphoplasia) mice that lacked all LN, PP, as well as newly identified intestinal isolated lymphoid follicles (ILF) (39). We confirmed that these two kinds of mice harbored numerous -IEL in the epithelial compartments of the small intestines and were found to retain gut CP. The significance of these findings is discussed from the viewpoint that all LN, PP, and ILF are dispensable anatomical sites for the generation of TI-IEL in the athymic nu/nu condition.
Materials and Methods
Mice
BALB/cA Jcl nu/nu (nu/nu), BALB/cA Jcl nu/+ (nu/+), aly/aly Jcl mutant and C.B-17/Icr Jcl scid/scid (scid/scid) mice were purchased from CLEA Japan (Tokyo, Japan). TCR- (KN6)-Tg mice on the BALB/c background were described previously (40). Female KN6-Tg mice were crossed with scid/scid mice to generate KN6 scid/+ mice, then they were backcrossed with scid/scid mice to obtain KN6 scid/scid mice. The presence of KN6-Tg was determined by PCR analysis of tail DNA with a set of primers to the KN6 Tg (5'-CAGATCCTTCCAGTTCATCC-3' and 5'-CAGTCACTTGGGTTCCTTGTCC-3'), and the homozygous scid/scid genotype was determined by the absence of TCR-+ T cells in PBL. We generated transplacentally manipulated nu/nu and nu/+ mice that lack all LN and PP according to essentially the same method described previously (38). In brief, timed-pregnant nu/+ mice that had been mated with male nu/nu mice were i.v. injected with 200 μg of both LT-R-Ig and TNFR-55-Ig fusion proteins on gestational days 13 and 16. All mice used for experiments were between 8 and 18 wk of age, and absence of a thymus in various athymic mice was checked at necropsy. All animal procedures described in this study were performed in accordance with the guidelines for animal experiments of Keio University School of Medicine.
Production of nu/nu aly/aly mice and genotyping of aly mutation
Because nu/nu and aly/aly mothers are incapable of nursing the neonates, we used an in vitro fertilization technique (41) to produce (nu/nuxaly/aly)F1 hybrid mice, then these heterozygous nu/+ aly/+ mice were intercrossed to obtain nu/nu aly/+, nu/nu aly/aly, and nu/+ aly/+ nu/+ aly/aly littermates. To determine aly/aly, aly/+, and +/+ alleles, the TaqMan assay of tail DNA was performed using the ABI PRISM 7000 sequence detection system (PerkinElmer) as previously described (42). The primer sequences for aly were 5'-GCCTACTGACATCCCGAGCTA-3' (forward primer) and 5'-GCAGGACTGGGCTGGAAGA-3' (reverse primer). The oligonucleotide probe corresponding mutant aly allele was 5'-AGACCGTACTGTTGAAG-3' (FAM labeled), and the oligonucleotide probe corresponding wild-type allele was 5'-AGACCGTACCGTTGAA-3' (VIC labeled). Underlining in sequences indicates point mutation. The 3' end of each probe carried the quencher that suppressed the fluorescence of the reporter dyes. Each DNA sample was amplified with the TaqMan Universal master mixture containing AmpliTaq Gold DNA polymerase according to the manufacturer’s instructions (Applied Biosystems). PCR conditions were 2 min at 50°C, 10 min at 95°C, 15 s at 95°C, and 1 min at 60°C for 40 cycles. During PCR, fluorescence developed when the oligonucleotide hybridized to perfectly matching DNA, and the exonuclease activity of Taq polymerase separated the quencher from the reporter dye. After PCR, the fluorescence yield for the two different dyes was measured and presented in a two-dimensional graph.
Antibodies
The following mAbs, described previously (22, 28, 29, 30, 39), were used. For immunohistochemical and immunofluorescence stainings: anti- (GL-3), anti-c-Kit (ACK-2), anti-B220 (RA3-6B2), and anti-IgA (C10-3) were used. For flow cytometric analysis, FITC-conjugated anti- (H57-597), anti- (GL-3), anti-V1 (2.11; gift from Dr. S. Tonegawa, Center for Learning and Memory, MIT, Cambridge, MA), anti-V4 (UC3-10A6), anti-V7 (GL-1; gift from Dr. L. Lefrancois, Department of Medicine, Division of Immunology, University of Connecticut Health Center, Farmington, CT), anti-CD4 (GK 1.5), biotinylated anti- (GL-3), anti-B220 (RA3-6B2), anti-CD8 (53-6.7), and anti-c-kit (ACK-2), and PE-conjugated anti-CD8 (53-5.8) and anti-CD4 (GK 1.5) were used.
Immunohistochemical procedure
Longitudinally opened small intestine, 10 mm in length, was pasted on a filter paper to form a horizontal section and then embedded in OCT compound (Tissue-Tek; Miles) at –80°C. The tissue segments were sectioned with a cryostat at 6 μm, and sections were preincubated with Block-Ace (Dainippon Pharmaceutical) to block nonspecific binding of mAbs. The sections were then incubated with hamster (anti-) or rat (anti-c-Kit, anti-B220, or anti-IgA) mAb for 30 min at 37°C and rinsed three times with PBS, followed by incubation with biotin-conjugated goat anti-hamster IgG Ab (5 μg/ml; Cedarlane Laboratories) or with biotin-conjugated goat anti-rat IgG (5 μg/ml; Cedarlane Laboratories). Subsequently, the sections were washed three times with PBS, then incubated with avidin-biotin peroxidase complexes (Vectastain ABC kit; Vector Laboratories). Histochemical color development was achieved with Vectastain 3,3'-diaminobenzidine substrate kit (Vector Laboratories) according to the manufacturer’s instructions. Finally, the sections were counterstained with hematoxylin for microscopy. Endogenous peroxidase activity was blocked with 0.3% H2O2 and 0.1% NaN3 in distilled water for 10 min at room temperature. Tissue sections incubated with either nonimmune hamster serum or isotype-matched normal rat IgG showed only minimal background staining.
Immunofluorescence procedure
Tissue segments from thymus, spleen, inguinal LN, MLN, and PP from KN6 scid/scid mice were embedded in OCT compound at –80°C. The small intestine of KN6 scid/scid mice was longitudinally opened along the mesenteric wall, then intestine, 10 mm in length, that had been rolled to form a vertical section was embedded in OCT compound at –80°C. Cryostat tissue sections, 6-μm thick, were fixed in acetone for 10 min at room temperature, washed three times with PBS, then pretreated with Block-Ace. Subsequently, the sections were incubated with anti-c-Kit mAb (ACK-2) for 60 min at 4°C, followed by incubation with PE-conjugated goat F(ab')2 anti-rat IgG (H+L) (Invitrogen Life Technologies). The sections were then incubated with anti- mAb (GL-3) and counterstained with FITC-conjugated goat anti-hamster IgG (H+L) (Jackson ImmunoResearch Laboratories). Finally, the sections were examined under a fluorescence microscope (Axiovert 100; Carl Zeiss) equipped with an image analysis system (Signal Analytics).
Flow cytometry
IEL were isolated according to methods described previously (22). Lymphoid cells were incubated first with biotinylated mAb, then with streptavidin-PE (BD Biosciences) and FITC-conjugated second mAb. Stained cells were suspended in staining medium (Hanks’ solution without phenol red, 0.02% NaN3, and 2% heat-inactivated FBS) containing 0.5 μg/ml propidium iodide and analyzed using FACScan with CellQuest software (BD Biosciences). Dead cells were excluded by propidium iodide gating. Three-color analysis of IEL was also performed. IEL were incubated first with anti-CD8 mAb (biotinylated), then with streptavidin-Tri-Color (Caltag Laboratories). After washing, IEL were counterstained with two combinations of two PE-conjugated mAbs (anti-CD8 and anti-CD4 mAbs) and one FITC-conjugated mAb (anti-), respectively. Lymphoid cells were incubated with anti-FcR II/III mAb (2.4G2) before staining to block nonspecific binding of labeled mAbs to FcR.
Results
Development of -IEL in nu/nu mice is independent of all LN and PP
To explore whether MLN are essential anatomical sites for the generation of TI-IEL in athymic nu/nu mice (37), we generated nu/nu mice that lacked all LN and PP. Pregnant nu/+ female mice that had been mated with nu/nu male mice were injected with LT-R-Ig and TNF-R55-Ig fusion proteins according to the protocol described by Rennert et al. (38), and the presence or the absence of LN and PP was determined in the progeny at 8 wk of age under a stereomicroscope. Although PP were absent from all treated mice, markedly attenuated remnants of MLN were present in about one-fifth of them. However, in every MLN-deficient nu/nu and nu/+ offspring, the development of mandibular, axillary, inguinal, and popliteal (data not shown) LN (i.e., peripheral LN) and cervical (data not shown), iliac, and sacral LN (i.e., mucosal LN) was also ablated (LNP– mice; Fig. 1).
FIGURE 1. Development of PP and all LN in athymic nu/nu mice is ablated by in utero treatment with both LT-R-Ig and TNF-R55-Ig fusion proteins. PP (A), MLN (B), mandibular LN (C), axillary LN (D), iliac LN (E), sacral LN (F), and inguinal LN (G) are present in untreated nu/nu mice (arrowheads), but are undetectable in nu/nu mice that were treated in utero with both LT-R-Ig and TNF-R55-Ig fusion proteins (nu/nu LNP– mice). Bar, 2 mm.
Flow cytometric analysis of IEL isolated from nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice was performed using anti-TCR- and anti-TCR- mAbs. Consistent with well-established findings (3, 4, 8, 43), the proportion of -IEL to -IEL was sharply reduced, and the composition of IEL not expressing either type of TCR was expanded in athymic nu/nu conditions regardless of the presence or the absence of all LN and PP (Fig. 2A). In contrast, no significant differences in absolute numbers of IEL were observed between LNP– and control LNP+ mice (data not shown). Notably, although the population size was slightly smaller by a factor of 1.5 compared with that in IEL from control nu/nu mice, a large number of -IEL was detected in IEL from nu/nu LNP– mice (Fig. 2A).
Compartmentalization of -IEL within the epithelial layer of small intestine in nu/nu LNP– mice was also verified by immunohistochemistry (Fig. 2B). These results indicate that the development of -IEL per se is independent of thymus, all LN, and PP.
FIGURE 2. Flow cytometric analysis of IEL and immunohistochemical examination of small intestines from nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice. A, Although the population size of -IEL in nu/nu LNP– mice that lack all LN and PP is smaller than that of nu/nu mice, a substantial number of -IEL are present in the epithelial compartment of nu/nu LNP– mice, indicating that -IEL are capable of developing in the absence of thymus, all LN including MLN, and PP. Note that the composition of -IEL is reduced drastically in the athymic nu/nu condition compared with that in the euthymic nu/+ condition. B, Representative immunohistochemical visualization of -IEL in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). Although numbers of -IEL in nu/nu LNP– mice are decreased compared with those in nu/nu mice, colonization of -IEL takes place in the absence of thymus, all LN including MLN, and PP (arrowheads). C, Representative immunohistochemical verification of CP in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). An average number and an approximate mass of CP filled with c-Kit+B200+ lymphocytes in these four different mice remain almost the same. D, Representative immunohistochemical verification of ILF in the small intestines of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice (magnification, x200). Note that a cluster of B220+ B cells that reside in the central region of ILF is surrounded by the layer of cells expressing c-Kit molecules.
With these findings in mind, we examined whether CP and ILF were present in these in utero manipulated LNP– mice, because, in contrast to PP that are already microscopically well developed just before birth (44), organogenesis of CP (28) and ILF (39) commences in early postnatal life. In fact, it was corroborated that the development of CP filled with closely packed c-Kit+ lymphocytes (Fig. 2C) and ILF containing B220+ B cell aggregation (Fig. 2D) remained intact in these nu/nu LNP– and nu/+ LNP– mice.
Development of -IEL in nu/nu aly/aly double-mutant mice
To ascertain the universality of the above findings, we explored the development of IEL in a mutant mouse that inherently lacked thymus, all LN, and PP, because fusion protein-treated LNP– mice might possess a minute and stereomicroscopically invisible MLN, even though this possibility appeared to be remote (38). With this purpose in mind, we generated nu/nu aly/aly double-mutant mice. In accordance with the earliest description (45), nu/nu aly/aly mice were devoid of all LN and PP (data not shown) as well as thymus. Importantly, substantial colonization of -IEL in the small intestine of nu/nu aly/aly mice was verified by flow cytometric (Fig. 3A) and immunohistochemical (Fig. 3B) analyses. Furthermore, as inferred from our previous observations (28, 39), histogenesis of CP was detected (Fig. 3C), whereas development of ILF was completely blocked (Fig. 3D), in these double-mutant animals. Taking all of these results together (Figs. 2 and 3), neither thymus, all LN including MLN, PP, nor ILF is an absolute requirement for the development of -IEL.
FIGURE 3. Flow cytometric analysis of IEL and immunohistochemical examination of small intestines from nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice. A, Although the population size of -IEL in nu/nu aly/aly mice that lack all LN, PP, and ILF is smaller than that in nu/nu aly/+ mice, a significant number of -IEL are present in the epithelial compartment of nu/nu aly/aly mice, indicating that -IEL are capable of developing in the absence of thymus, all LN including MLN, PP, and ILF. Note that the composition of -IEL is reduced drastically in the athymic nu/nu condition compared with that in the euthymic nu/+ condition. B, Representative immunohistochemical verification of -IEL in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Although numbers of -IEL in nu/nu aly/aly mice are lower by a factor of 1.5–2 compared with those in nu/nu aly/+ mice, a significant colonization of -IEL takes place in the absence of thymus, all LN including MLN, PP, and ILF (arrowheads). C, Representative immunohistochemical verification of CP in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Although an average number of CP filled with c-Kit+B220– lymphocytes in these four different mice remain almost the same, an approximate mass of CP present in nu/nu aly/aly and nu/+ aly/aly mice is slightly reduced compared with that of CP present in nu/nu aly/+ and nu/+ aly/+ mice. D, Representative immunohistochemical verification of ILF in the small intestines of nu/nu aly/+, nu/nu aly/aly, nu/+ aly/+, and nu/+ aly/aly mice (magnification, x200). Note that ILF are undetectable throughout the small intestinal mucosa of nu/nu aly/aly and nu/+ aly/aly mice.
In this context, it is important to determine T and B cells and IgA+ B cells that sojourn, respectively, in the spleen and LP of nu/nu LNP– and nu/nu aly/aly mice, because spleen is most likely the sole organized peripheral lymphoid tissue remaining in these animals, and nu/nu aly/aly mice lack intestinal IgA+ B cell-producing plants such as PP and ILF. In contrast to abundant B220+ B cells, mature T cells were virtually absent in the spleens of nu/nu LNP– (Fig. 4A) and nu/nu aly/aly (Fig. 4C) mice. In the small intestines, however, nu/nu LNP– mice possessed IgA+ B cells (Fig. 4B), -IEL, CP, and well-developed ILF (Fig. 2), whereas nu/nu aly/aly mice possessed -IEL (Fig. 3, A and B) and CP (Fig. 3C), but lacked IgA+ B cells (Fig. 4C) and ILF (Fig. 4D). These results indicate that all LN, including MLN, PP, and ILF, are not an absolute requirement for the generation of -IEL in nu/nu mice and that the aly mutation interferes the formation of PP and ILF, resulting in the impaired development of IgA+ B cells in villous LP.
FIGURE 4. Flow cytometric analysis of splenic T and B cells and immunohistochemical verification of intestinal IgA+ B cells in nu/nu, nu/nu LNP–, nu/nu aly/+, and nu/nu aly/aly mice. A, Although mature CD3+ T cells are nearly absent from the spleens of nu/nu and nu/nu LNP– mice, abundant B220+ B cells are detected in the spleens of these athymic animals. Colonization of IgA+ B cells in the villous LP of nu/nu, nu/nu LNP–, nu/+, and nu/+ LNP– mice is comparable. B, Only a marginal number of mature CD3+ T cells are detected in the spleens of nu/nu aly/+ and nu/nu aly/aly mice, and a large fraction of the remaining lymphoid cells consists of B220+ B cells. In contrast, villous LP of nu/nu aly/aly and nu/+ aly/aly mice has no IgA+ B cells.
Phenotypic and V gene usage analyses of -IEL in nu/nu LNP– and nu/nu aly/aly mice
The data reported to date indicate that -IEL are potentially capable of developing in nu/nu mice lacking all LN, PP, and ILF. In this regard, however, it is reasonable to consider the possibility that -IEL generated under such harsh conditions might differ from those generated in nu/nu mice possessing all LN, PP, and ILF. To address this issue, we conducted flow cytometric analysis of -IEL isolated from nu/nu, nu/nu LNP–, and nu/nu aly/aly mice. Although absolute numbers of -IEL were lower by a factor of 2 compared with those in nu/nu mice (Fig. 5A), the composition of the major -IEL subset expressing CD8 homodimer (Fig. 5A) and the V1, V4, and V7 gene segment used by such -IEL (Fig. 5B) remained the same in both nu/nu LNP– and nu/nu aly/aly mice. Collectively, these results indicate that the absence of all LN, PP, and ILF exerts only a small effect on the phenotypic configuration of -IEL in nu/nu mice.
FIGURE 5. Flow cytometric analysis of -IEL from nu/nu, nu/nu LNP–, nu/nu aly/+, and nu/nu aly/aly mice and V gene segments used by -IEL in these athymic nu/nu mice. A, Three-color analysis was performed. Absolute numbers of double-negative (CD4–8–), single-positive (CD4+, CD8+, or CD8+), and double-positive (CD4+8+) subsets in the -IEL population were calculated on the basis of total numbers of -IEL. Data are the mean values from five mice per group. B, Two-color analysis was performed. IEL isolated from these nu/nu mice were incubated first with anti-TCR- mAb (biotinylated), then with streptavidin-PE and FITC-conjugated anti-V1, anti-V4, or anti-V7 mAb. The results are the mean ± SD of data obtained from four mice per group.
Development of c-Kit+ -IEL in scid/scid mice expressing Tg TCR-
We have shown that IL-7 produced by intestinal epithelial cells (IEC) is important for intraintestinal development of -IEL and is crucial for organization of intestinal mucosal lymphoid tissues, such as PP and CP (18). It has also been shown that c-Kit and stem cell factor (SCF) are expressed by -IEL and IEC, respectively, and signaling through c-Kit/SCF is indispensable for normal development of -IEL (46, 47). Thus, these previous (46, 47) and the present findings in conjunction with results obtained with BM chimeric mice (30), reinforce the idea that the development of -IEL takes place in the intestinal mucosa in situ. In an attempt to confirm and visualize directly the cellular events that proceed toward gut-oriented -IEL generation, i.e., c-Kit+ CP cellsc-Kit+ TCR- T cellsc-Kit– -IEL, we generated scid/scid mice expressing KN6-Tg TCR- (48). Double-immunofluorescence analysis of small intestinal tissues containing CP highlighted these presumptive cellular events. Thus, a representative picture of jejunal tissue sections from KN6 scid/scid mice (Fig. 6A) shows that a cluster of c-Kit+ cells in a CP (red) does not express TCR-, and that large numbers of -IELs (green) are present in the epithelial layer, especially in the epithelium adjacent to CP. Notably, quite a large number of lymphocytes expressing both c-Kit and TCR- molecules (yellow or orange) is also present in the epithelial and LP compartments of villi (Fig. 6A). Neither clustering c-Kit+ cells nor lymphocytes stained yellow/orange were detected in other lymphoid tissues, such as thymus, LN, PP, and spleen (data not shown). Flow cytometric analysis of cell surface c-Kit molecules on the gated T cells confirmed that a large fraction of -IEL was c-Kit positive, namely, double-positive TCR-+c-Kit+ cells (Fig. 6B, upper panel). In contrast, abundant Tg T cells residing in MLN, spleen, and thymus did not include such c-Kit-expressing, double-positive cells (Fig. 6B, lower three panels). Overall, these results support the above-described basic premise that the development of -IEL takes place in the intestinal mucosa in situ. However, because KN6-Tg TCR--expressing cells have not been detected in the CP of KN6 scid/scid mice (Fig. 6A), whether CP are essential and indispensable anatomical sites in generating -IEL remained highly contentious.
FIGURE 6. Double immunofluorescence analysis of small intestinal tissues containing CP and flow cytometric analysis of lymphocytes from KN6 scid/scid mice. A, Intestinal birthplace of intraepithelial T cells (magnification, x400). A cluster of lymphocytes in a CP is positively stained with anti-c-Kit mAb (PE), but not with anti- mAb (FITC), resulting in a red color. Conversely, many IEL, especially those adjacent to CP, are positively stained with anti- mAb (FITC), but not with anti-c-Kit mAb (PE), resulting in a green color. Note that a large number of lymphocytes present in the epithelial and LP compartments of villi are positively stained with anti-c-Kit mAb (PE) and anti- mAb (FITC), resulting in a yellow or orange color. Neither the clusters filled with c-Kit+ cells nor lymphocytes stained yellow or orange are detected in the other lymphoid organs, such as thymus, LN, and spleen. B, Expression of c-Kit molecules by Tg T cells that colonize in the intestinal epithelial, MLN, and splenic and thymic compartments of KN6 scid/scid mice. IEL, MLN cells, spleen cells, and thymocytes were incubated first with anti-TCR- mAb (biotinylated), then with streptavidin-PE and FITC-conjugated anti-c-Kit mAb. Profiles of c-Kit expression by the gated TCR-+ cell population are presented. The dark area in the upper panel depicts a FITC-positive cell profile of IEL that were stained simply with biotinylated anti-TCR- mAb and streptavidin-PE (negative control). It is evident that IEL contain a large number of double-positive (TCR-+ c-Kit+) cells, but that MLN, spleen, and thymus have no such double-positive cells.
Discussion
Because most of the numerous lymphocytes residing in the murine IEC compartment unexpectedly turned out to be T cells (IEL), additional revelation toward the end of the last century indicated that they display phenotypic and functional characteristics distinct from those of other T cell populations and that their development does not necessarily depend on the thymus (TI-IEL) (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18). Likewise, several lines of information have illuminated the distinctive T cell facets of human fetal intestine (49, 50), and on the basis of RAG expression, there may also be TI-IEL that develop in human (51, 52) and rat (53) intestines. Although evidence for the vestigial lymphocyte-producing function of gut mucosa is substantial, as mentioned above (2, 5, 7, 11, 12, 18, 19, 20, 21, 22, 23, 24, 25, 49, 50, 51, 52, 53), recent studies have made it clear that CD8 does not serve as a marker for TI development of -IEL (26, 27). In contrast, because positive and/or negative selection of TCR- T cells in the thymus is not as evident (26, 27), thymic dependency for functional TCR- T cells is less obvious.
In an assessment of the expression of Tg-encoded GFP in place of RAG-2 protein, no evidence was obtained for a lymphopoietic process involving CP cells migrating into gut epithelium to undergo TCR gene rearrangement and maturation into - and -IEL even in the athymic nu/nu condition (37). Instead, MLN and, less efficiently, PP have been identified as the major extrathymic T cell-producing plants in athymic nu/nu mice, and the newly generated T cells migrate from MLN into thoracic duct lymph to reach the gut epithelia, indicating that MLN and PP are the pivotal sites in generating TI-IEL, mostly -IEL (37). In this context, the development of -IEL should be hampered in nu/nu mice that simultaneously lack MLN and PP. Our present findings argue against this scenario by showing that not only transplacentally manipulated nu/nu LNP– mice lacking all LN and PP, but also genetically defined nu/nu aly/aly mice lacking all LN, PP, and ILF harbor a substantial population of -IEL (Fig. 2, A and B, and Fig. 3, A and B), although these two different mouse models may share the same confounding factors. In contrast to what was observed in nu/nu mice, the extrathymic pathway of IEL generation was shown to be totally repressed in the euthymic condition using the same GFP RAG-2 Tg mouse model (37). The authors proposed that all IEL, including CD8+ IEL, in normal mice were the likely progeny of double-negative (DN) TCR-+ and -+ thymocytes and noted that extremely complex and unusual T cell characteristics of murine IEL with respect to their expression of various accessory, costimulation, activation, and adhesion markers (4, 5) might be brought about by the distinctive microenvironment of gut epithelium (37). In this context, for instance, many DN thymocytes somehow down-regulate cell surface expression of Thy-1 molecules, because a substantial fraction of IEL is Thy-1 negative, whereas most of them must up-regulate c-Kit molecules on the way to becoming IEL (46, 47) (Fig. 6). Even if all of those transfigurations (>20) are attributable to the inherent properties of gut epithelium, the biological significance as well as the molecular level of the mechanisms underlying such enigmatic cellular events remain highly contentious. It should also be pointed out that our recent findings (18) are inconsistent with this idea (37). T cells are absent in IL-7–/– mice due to the selective blockade of TCR- gene rearrangements (54). Using the intestinal fatty acid-binding protein (iFABP) promoter, we reinstated the expression of IL-7 to mature IEC of IL-7–/– mice (iFABP-IL7) (18). Although -IEL were restored in iFABP-IL7 mice as well as CP and PP, T cells remained absent from all tissues, including thymus, spleen, and skin. These results clearly indicate that -IEL generated in iFABP-IL7 mice are not of DN TCR-+ thymocyte origin and that the recombination of TCR- genes in TCR-precursor T cells takes place in situ with the assistance of IL-7 produced locally by IEC.
Our present findings provide compelling evidence for the development of -IEL within the intestine of nu/nu mice that lack the thymus, all LN, PP, and ILF. It should be pointed out, however, that the population size and absolute numbers of -IEL from nu/nu LNP– and nu/nu aly/aly mice are smaller than those from the corresponding control nu/nu mice (Figs. 2A, 3A, and 5A). These features would indicate that LN and PP, in fact, contribute to -IEL numbers in the nu/nu condition. In contrast to the results obtained in GFP RAG-2 Tg mouse model (37), our previous RT-PCR analysis of lymphocytes from normal euthymic mice showed that under conditions in which mRNA from 50 thymocytes displayed a strong signal for RAG-2 transcripts and mRNA from 6250 RAG-2–/– thymocytes failed to display any detectable signals, very low levels of RAG-2 transcripts were constantly detected in an amount of mRNA equivalent to 6250 IEL and CP cells (22). These findings suggest that a small minority of IEL and possibly CP cells also is undergoing TCR gene rearrangement, that they are able to do so with a minimum amount of RAG-2 transcripts, or both. Actually, we still do not know how many RAG-1 and -2 molecules per nucleus are required to drive the region-specific V(D)J recombinations of TCR genes. It is possible that the amount of mRNA encoding RAG-1 and -2 molecules required by thymocytes for the successful recombination of TCR genes may not be that large. It should also be pointed out that using cell fate mapping, almost all -IEL have recently been shown to be the progeny of immature CD4+CD8+ thymocytes (55). Furthermore, by using elegant and sophisticated approaches (55), it has been revealed that the retinoic acid-related orphan receptors (RORt) detected in fetal lymphoid tissue-inducer cells are also expressed in cells within gut CP, and that -IEL are not the progeny of such RORt-positive CP lymphocytes. Specifically, however, it has remained an open question whether a small, but significant, fraction of lymphocytes in CP does not retain RORt molecules or whether almost all CP lymphocytes express RORt (55). In any event, to substantiate the intraintestinal development of -IEL in these nu/nu LNP– and nu/nu aly/aly mice, clear identification of T cells undergoing TCR gene rearrangement in the gut mucosa appears to be of critical importance.
Acknowledgments
We are grateful to Dr. L. Lefrancois for his critical reading of the manuscript. We thank N. Hosaka and M. Mori for their excellent technical assistance.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Grant-in-Aid for Creative Scientific Research (13GS0015); the Japan Society for the Promotion of Science; a Grant-in-Aid for Scientific Research on Priority Areas A; a Grant-in-Aid for the 21st Century of Excellence Program entitled Understanding and Control of Life’s Function via Systems Biology (Keio University); the Special Coordination Fund for Promoting Science and Technology, Ministry of Education, Culture, Sports, Science, and Technology; and Health Science Research Grants from the Ministry of Health, Labor, and Welfare.
2 Address correspondence and reprint requests to Dr. Hiromichi Ishikawa, Department of Microbiology and Immunology, Keio University School of Medicine, Shinjuku-ku, Tokyo 160-8582, Japan. E-mail address: h-ishika@sc.itc.keio.ac.jp
3 Abbreviations used in this paper: IEL, intestinal intraepithelial T cell; aly, alymphoplasia; BM, bone marrow; CP, cryptopatch; DN, double negative; iFABP, intestinal fatty acid-binding protein; IEC, intestinal epithelial cell; ILF, isolated lymphoid follicle; Lin, lineage marker; LNP–, lymph node- and Peyer’s patch deficient; LP, lamina propria; LT, lymphotoxin; LT-R, lymphotoxin- receptor; MLN, mesenteric lymph node; PP, Peyer’s patch; RORt, retinoic acid-related orphan receptor; SCF, stem cell factor; Tg, transgenic; TI, thymus independent.
Received for publication August 9, 2004. Accepted for publication December 10, 2004.
References
Rocha, B., D. Guy-Grand, P. Vassalli. 1995. Extrathymic T cell differentiation. Curr. Opin. Immunol. 7:235.
Mowat, A. M., J. L. Viney. 1997. The anatomical basis of intestinal immunity. Immunol. Rev. 156:145.
Klein, J. R.. 1998. Thymus-independent development of gut T cells. Chem. Immunol. 71:88.
Lefrancois, L., L. Puddington. 1999. Basic aspects of intraepithelial lymphocyte immunobiology. P. L. Ogra, and J. Mestecky, and M. E. Lamm, and W. Strober, and J. Bienenstock, and J. R. McGhee, eds. Mucosal Immunology 413. Academic Press, San Diego.
Aranda, R., B. C. Sydora, M. Kronenberg. 1999. Intraepithelial lymphocytes: function. P. L. Ogra, and J. Mestecky, and M. E. Lamm, and W. Strober, and J. Bienenstock, and J. R. McGhee, eds. Mucosal Immunology 429. Academic Press, San Diego.
Poussier, P., M. Julius. 1999. Speculation on the lineage relationships among CD4–8+ gut-derived T cells and their role(s). Semin. Immunol. 11:293.
Hayday, A., E. Theodoridis, E. Ramsburg, J. Shires. 2001. Intraepithelial lymphocytes: exploring the Third Way in immunology. Nat. Immunol. 2:997.
Guy-Grand, D., P. Vassalli. 2002. Gut intraepithelial lymphocyte development. Curr. Opin. Immunol. 14:255.
Goodman, T., L. Lefrancois. 1988. Expression of the - T-cell receptor on intestinal CD8+ intraepithelial lymphocytes. Nature 333:855.
Bonneville, M., C. A. Janeway, K. Ito, W. Haser, I. Ishida, N. Nakanishi, S. Tonegawa. 1988. Intestinal intraepithelial lymphocytes are a distinct set of T-cells. Nature 336:479.
Lefrancois, L.. 1991. Phenotype complexity of intraepithelial lymphocyte of the small intestine. J. Immunol. 147:1746.
Guy-Grand, D., N. Cerf-Bensussan, B. Malissen, M. Malassis-Seris, C. Briottet, P. Vassalli. 1991. Two gut intraepithelial CD8+ lymphocyte populations with different T cell receptors: a role for the gut epithelium in T cell differentiation. J. Exp. Med. 173:471.
Maloy, K. J., A. M. Mowat, R. Zamoyska, I. N. Crispe. 1991. Phenotypic heterogeneity of intraepithelial T lymphocytes from mouse small intestine. Immunology 72:555
Kawaguchi, M., M. Nanno, Y. Umesaki, S. Matsumoto, Y. Okada, Z. Cai, T. Shimamura, Y. Matsuoka, M. Ohwaki, H. Ishikawa. 1993. Cytolytic activity of intestinal intraepithelial lymphocytes in germ-free mice is strain dependent and determined by T cells expressing T-cell antigen receptors. Proc. Natl. Acad. Sci. USA 90:8591.
Malissen, M., A. Gillet, B. Rocha, J. Trucy, E. Vivier, C. Boyer, F. Kontgen, N. Brun, G. Mazza, E. Spanopoulou, et al 1993. T cell development in mice lacking CD3-/ gene. EMBO J. 12:4347.
Ohno, H., T. Aoe, S. Taki, D. Kitamura, Y. Ishida, K. Rajewsky, T. Saito. 1993. Development and functional impairment of T cells in mice lacking CD3 chains. EMBO J. 12:4357.
Liu, C-P., R. Ueda, J. She, J. Sancho, B. Wang, G. Weddell, J. Loring, C. Kurahara, E. C. Dudley, A. Hayday, et al 1993. Abnormal T cell development in CD3–/– mutant mice and identification of a novel T cell population in the intestine. EMBO J. 12:4863
Laky, K., L. Lefrancois, E. G. Lingenheld, H. Ishikawa, J. M. Lewis, S. Olson, K. Suzuki, R. E. Tigelaar, L. Puddington. 2000. Enterocyte expression of IL-7 induces development of T cells and Peyer’s patches. J. Exp. Med. 191:1569.
Guy-Grand, D., C. V. Broecke, C. Briottet, M. Malassis-Seris, F. Selz, P. Vassalli. 1992. Different expression of the recombination activity gene RAG-1 in various populations of thymocytes, peripheral T cells and gut thymus-independent intraepithelial lymphocytes suggests two pathways of T cell receptor rearrangement. Eur. J. Immunol. 22:505.
Lin, T., G. Matsuzaki, H. Yoshida, N. Kobayashi, H. Kenai, K. Omoto, K. Nomoto. 1994. CD3–CD8+ intestinal intraepithelial lymphocytes (IEL) and the extrathymic development of IEL. Eur. J. Immunol. 24:1080.
Boll, G., A. Rudolphi, S. Spieb, J. Reimann. 1995. Regional specialization of intraepithelial T cells in the murine small and large intestine. Scand. J. Immunol. 41:103.
Oida, T., K. Suzuki, M. Nanno, Y. Kanamori, H. Saito, E. Kubota, S. Kato, M. Itoh, S. Kaminogawa, H. Ishikawa. 2000. Role of gut cryptopatches in early extrathymic maturation of intestinal intraepithelial T cells. J. Immunol. 164:3616.
Hamad, M., M. Whetsell, J. Wang, J. R. Klein. 1997. T cell progenitors in the murine small intestine. Dev. Comp. Immunol. 21:435.
Page, S. T., L. Y. Bogatzki, J. A. Hamerman, C. H. Sweenie, P. J. Hogarth, M. Malissen, R. M. Perlmutter, A. M. Pullen. 1998. Intestinal intraepithelial lymphocytes include precursors committed to the T cell receptor lineage. Proc. Natl. Acad. Sci. USA 95:9459.
Woodward, J., E. Jenkinson. 2001. Identification and characterization of lymphoid precursors in the murine intestinal epithelium. Eur. J. Immunol. 31:3329.
Leishman, A. J., L. Gapin, M. Capone, E. Palmer, H. R. MacDonald, M. Kronenberg, H. Cheroutre. 2002. Precursors of functional MHC class I- or class II-restricted CD8+ T cells are positively selected in the thymus by agonist self-peptides. Immunity 16:355.
Cheroutre, H.. 2004. Starting at the beginning: new perspectives on the biology of mucosal T cells. Annu. Rev. Immunol. 22:217.
Kanamori, Y., K. Ishimaru, M. Nanno, K. Maki, K. Ikuta, H. Nariuchi, H. Ishikawa. 1996. Identification of novel lymphoid tissues in murine intestinal mucosa where clusters of c-kit+ IL-7R+ Thy1+ lympho-hemopoietic progenitors develop. J. Exp. Med. 184:1449.
Saito, H., Y. Kanamori, T. Takemori, H. Nariuchi, E. Kubota, H. Takahashi-Iwanaga, T. Iwanaga, H. Ishikawa. 1998. Generation of intestinal T cells from progenitors residing in gut cryptopatches. Science 280:275.
Suzuki, K., T. Oida, H. Hamada, O. Hitotsumatsu, M. Watanabe, T. Hibi, H. Yamamoto, E. Kubota, S. Kaminogawa, H. Ishikawa. 2000. Gut cryptopatches: direct evidence of extrathymic anatomical sites for intestinal T lymphopoiesis. Immunity 13:691.
Alt, F. W., T. K. Blackwell, G. D. Yancopoulos. 1987. Development of the primary antibody repertoire. Science 238:1079.
Hempel, W. M., I. Leduc, N. Mathieu, R. K. Tripathi, P. Ferrier. 1998. Accessibility control of V(D)J recombination: lessons from gene targeting. Adv. Immunol. 69:309.
Ye, S.-K., K. Maki, T. Kitamura, S. Sunaga, K. Akashi, J. Domen, I. L. Weissman, T. Honjo, K. Ikuta. 1999. Induction of germline transcription in the TCR locus by Stat5: implications for accessibility control by the IL-7 receptor. Immunity 11:213.
Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H. J. Fehling, H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science 266:1208.
Wilson, A., H. R. MacDonald. 1995. Expression of genes encoding the pre-TCR and CD3 complex during thymus development. Int. Immunol. 7:1659.[(Satoshi Nonaka, Tomoaki N)