HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hara, H. Articles by Nomoto, K. Search for Related Content PubMed PubMed Citation Articles by Hara, H. Articles by Nomoto, K.

Development of Dendritic Epidermal T Cells with a Skewed Diversity of TCRs in V1-Deficient Mice¹

Hiromitsu Hara^*, Kenji Kishihara^3,*, Goro Matsuzaki^*, Hiroaki Takimoto^2,*, Tadasuke Tsukiyama, Robert E. Tigelaar and Kikuo Nomoto^*

Departments of ^* Immunology and Molecular and Cellular Biology, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan; and Department of Dermatology, Section of Immunobiology, Yale Skin Diseases Research Core Center, Yale University, New Haven, CT 06520

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
One of the most intriguing features of T cells that reside in murine epithelia is the association of a specific V/V usage with each epithelial tissue. Dendritic epidermal T cells (DETCs) in the murine epidermis, are predominantly derived from the "first wave" V5⁺ fetal thymocytes and overwhelmingly express the canonical V5/V1-TCRs lacking junctional diversity. Targeted disruption of the V1 gene resulted in a markedly impaired development of V5⁺ fetal thymocytes as precursors of DETCs; however, TCR⁺ DETCs with a typical dendritic morphology were observed in V1^-/- mice and their cell densities in the epidermis were slightly lower than those in V1^+/- epidermis. Moreover, the V1-deficient DETCs were functionally competent in their ability to up-regulate cytokines and keratinocyte growth factor-expression in response to keratinocytes. V5⁺ DETCs were predominant in the V1^-/- epidermis, though V5^- TCR⁺ DETCs were also detected. The V5⁺ DETCs showed a typical dendritic shape, TCR^high, and age-associated expansion in epidermis as observed in conventional DETCs of normal mice, whereas the V5^- TCR⁺ DETCs showed a less dendritic shape, TCR^low, and no expansion in the epidermis, consistent with their immaturity. These results suggest that optimal DETC development does not require a particular V/V-chain usage but requires expression of a limited diversity of TCRs, which allow DETC precursors to mature and expand within the epidermal microenvironment.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
An ontogenic feature of murine fetal thymocyte development is the ordered and overlapping appearance of waves of cells expressing a TCR composed of specific V- and V-chains (1, 2, 3, 4, 5). The first two waves in thymic ontogeny are unique in that they express invariant TCRs characterized by a lack of junctional diversity. The first wave appearing around gestation day (GD)⁴ 14–16 expresses a canonical V5 (GV1 by World Health Organization-International Union of Immunological Sciences nomenclature, see Ref. 6)-J1-C1 chain, and the second wave appearing around GD16–18 expresses a canonical V6 (GV2)-J1-C1 chain. Both the V5 and V6 chains preferentially pair with a common canonical V1 (DV101)-D2-J2-C chain (5, 7, 8, 9). These V5/V1 and V6/V1 subsets home to epidermis and to mucosal epithelia of reproductive tract and tongue, respectively (7, 10). Later, T subsets expressing variant V1 (GV5S1), V4 (GV3), and V7 (GV4) chains in combination with multiple V-chains appear in neonatal and adult thymus and home to blood and peripheral lymphoid organs (1, 4, 11), as well as into other epithelial tissues such as the gastrointestinal tract (12, 13, 14).

Although the vast majority of T cells in the peripheral blood, lymph nodes, and spleen of adult mice express ßTCRs, T cells are a major T cell population in murine epithelia interfacing with external environment (e.g., skin, reproductive tract, gastrointestinal tract, and lung; Refs. 7, 11, 12 , and 15). One of the most intriguing features of these T cells is the association of a specific V/V usage with each epithelial tissue. Particularly, in the epithelia of skin and the reproductive tract, there are unique populations of T cells with a highly restricted TCR repertoire as described above. Virtually all the reproductive tract epithelial T cells express the canonical V6/V1-TCR, whereas the vast majority of the T cells in the epidermis, referred to as dendritic epidermal T cells (DETCs), express the canonical V5/V1-TCR (16, 17, 18). The presence of such identical TCRs on T cells in these sites is thought to be related to the capacity of such cells to recognize an as yet identified stress-induced autologous Ag; DETCs have been reported to be stimulated in vitro by contact with stressed keratinocytes or a transformed keratinocyte line in a MHC-nonrestricted manner (16, 19). Production of various cytokines by DETCs after in vitro and in vivo activation and non-MHC-restricted cytotoxicity of DETC lines against tumor targets resembling lymphokine-activated killer cell activity has also been demonstrated (20, 21, 22, 23, 24, 25), consistent with a role for DETCs as effector cells in immune surveillance of the skin. In addition, it has been shown that activated V5⁺ DETCs produce keratinocyte growth factor (KGF), consistent with a potential role for DETCs in wound healing (26, 27).

Transgenic TCR-expressing T cells were found in the epidermis of KN6 (V4/V5 (DV105)-TCR)-transgenic mice, suggesting that homing specificity of DETCs is independent of the TCR specificity (28). However, DETCs in the transgenic mice were smaller and were present in lower numbers/density than DETCs observed in normal mice. This observation was consistent with the possibility that a specific TCR usage might be required for complete development and/or maintenance of DETCs in epidermis. Recently, mice in which the V5 gene was disrupted by gene targeting were generated (29); the morphology, density, and functional activity of the TCR⁺ DETCs in such knockout mice was indistinguishable from that seen in littermate controls. A readily detectable, but variable proportion of the DETCs in the epidermis of V5-deficient mice expressed a TCR in which V1-J4-C4 chain was paired with a canonical V1-D2-J2-C chain. Furthermore, such V1/V1 DETCs could be stained with a mAb, 17D1, previously felt to recognize only the canonical V5/V1-TCR expressed by normal DETCs (30), suggesting that a specific TCR conformation rather than a simple linear epitope(s) composed of particular V- and V-chains may be required for the normal development and maintenance of DETCs in epidermis.

In this study, we generated V1 gene-deficient (V1^-/-) mice using a Cre/loxP gene targeting strategy (31, 32) in an attempt to clarify the requirement of V1 gene expression for the development of fetal thymocyte and of DETCs. In the V1^-/- mice, we found markedly impaired development of V5⁺ fetal thymocytes, which have been demonstrated to be DETC precursors; however, the DETCs in such mice have relatively normal morphology, density, and functional activity. The DETCs from V1^-/- mice predominantly expressed the canonical V5 chain paired with a fetal-type V6 (DV7, ADV7)-D2-J1-C chain with a limited junction, which was consistent with the possibility that they were derived from fetal thymocytes equally to normal DETCs. The V5⁺ DETCs, but not V5^- DETCs, in V1^-/- mice expressed a high level of TCRs and proliferate within the epidermis comparable to conventional V5⁺ DETCs in control mice. These results suggest that optimal DETC development requires expression of a limited diversity of TCRs, which allow DETC precursors to mature and expand within the epidermal microenvironment. In this report, the significance of TCR conformation in the development of DETCs is discussed.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Generation of V1 gene-deficient mice

An 11-kb mouse genomic DNA (BglII-BglII) fragment containing the V1 gene segments was obtained from the 129/SvJ mouse genomic library (Fig. 1A). A 0.9-kb DNA fragment between the 5'-flanking SacI site of the first exon and the EcoRV site located nearly at the 3' terminus of the second exon in the V1 gene was replaced with the loxP-flanked PGK-neo cassette. The final targeting construct consisted of a 0.9-kb short arm and a 6.3-kb long arm of homology linked by the loxP-flanked PGK-neo cassette. The targeting vector DNA linearized by SalI was electroporated into E14K embryonic stem (ES) cells, followed by selection in the presence of 300 µg/ml G418 for 7 days. ES cell clones with the expected homologous recombination event were screened by PCR and by Southern blot analysis of PstI-digested genomic DNA hybridized with a [³²P]-labeled DNA fragment 3' flanking the targeting vector. To excise the PGK-neo cassette from the V1-targeted allele in the ES clones, 25 µg supercoiled Cre-encoding plasmid DNA (pMCcre-puro vector, constructed and kindly provided by Dr. J. Takeda, Osaka University, Suita, Osaka, Japan; cre gene was supplied by Dr. T. Yagi, National Institute of Physiological Sciences, Okazaki, Aichi, Japan) were transferred into 10⁷ targeted ES cells by electroporation. After selection in 1 µg/ml puromycin-containing medium for 6–7 days, surviving ES colonies were isolated and the deletion event was identified by Southern blot analysis. ES clones with complete deletion of the PGK-neo cassette were microinjected into fertilized blastocysts from C57BL/6 mice. The resultant chimeric male mice with a high chimerity were crossed with C57BL/6 females to generate agouti mice with germline transmission. Following mating of heterozygously mutated mice, homozygotes were identified and distinguished from heterozygous and wild-type mice by Southern blot analysis.

View larger version (23K): [in this window] [in a new window]   FIGURE 1. Generation of V1-deficient mice. A, Endogenous V1 gene locus and targeting construct (SI, SacI; RI, EcoRI; RV, EcoRV; PI, PstI; XI, XhoI; and LI, SalI). A 0.9-kb SacI-EcoRV fragment containing 2 exons encoding the V1 gene (solid boxes) was replaced with loxP(closed arrow heads)-flanked PGKneo cassette (open arrow, NEO) by homologous recombination. The direction of the NEO insertion corresponds to that of the open arrow. NEO was removed by transient expression of Cre recombinase (Cre). ES clones that survived each selection were screened by Southern blot analysis of PstI-digested genomic DNAs, using a genomic DNA fragment flanked as a probe displayed in the figure (thick bar). B, Southern blot analysis of ES clones and the V1-deficient mice. Genomic DNA was extracted from the parental E14K ES line (lane 1), ES clone 5C with the V1-targeted allele (lane 2), ES clone 5C.5 with the NEO-deleted allele (lane 3), and tails of V1+/+ mice (lane 4), V1+/- mice (lane 5) and V1-/- mice (lane 6) littermates. Genomic DNA samples were digested with PstI and hybridized with a probe indicated in A. The V1-targeted allele and the NEO-deleted allele were detected as a 1.8-kb and 4.3-kb band, respectively, in comparison to the wild-type allele (5.2-kb band).

Epidermal cells were prepared as described previously (33). Briefly, both ventral and dorsal aspects of ear skin were separated from underlying cartilage using fine forceps, followed by flotation, dermal side up, on 1% trypsin/PBS solution for 45 min at 37°C. The epidermis was then separated from dermal tissue and epidermal single cell suspensions were prepared by mechanical agitation. The resulting epidermal cell suspensions were enriched and separated from dead cells by Lympholyte-M (Cedarlane Laboratories, Hornby, Ontario, Canada) density gradient centrifugation. Interface epidermal cells (IECs) were collected and cultured overnight in complete RPMI 1640 medium containing 10% FBS in the presence of 5U/ml human rIL-2 (R&D Systems, Minneapolis, MN). Before flow cytometric analysis, the cultured cells were separated from dead cells by centrifugation over Lympholyte-M.

In some experiments, IECs from V1^-/- mice were separated into V5⁺ cells and V5^- cells using a magnetic cell sorting system (Vario MACS; Miltenyi Biotec, Auburn, CA). Briefly, IECs were stained with FITC-conjugated anti-V5 (F536; PharMingen, San Diego, CA). After washing, the cells were labeled magnetically with anti-FITC Multisort Microbeads (Miltenyi Biotec). Labeled cells were applied onto a separation column (Type RS⁺; Miltenyi Biotec) placed in a magnetic field (Vario MACS), and the column was washed with PBS/BSA. The flow-through was collected as the V5^- cell fraction. Bound cells were eluted from the column with PBS/BSA after removing it from the magnetic field and were used as V5⁺ cells. The purity of each V5⁺ or V5^- fraction among CD3⁺ T cells was >95%.

Flow cytometric analysis

For four-color flow cytometric analysis, cells were stained with FITC-conjugated, PE-conjugated, allophycocyanin (APC)-conjugated, or biotin-conjugated mAb, followed by staining them with RED670-conjugated streptavidin (Life Technologies, Gaithersburg, MD). In this study, the following dye-conjugated mAbs were used: FITC-conjugated anti-C (GL3; PharMingen), anti-V5 (F536; PharMingen), anti-V1 (2.11; generously provided by Dr. P. Pereira, Institut Pasteur, Paris, France), anti-V5/V1 (17D1) mAbs, PE-conjugated anti-C (GL3; PharMingen) mAbs; biotin-conjugated anti-V5 (F536; PharMingen), anti-Thy-1.2 (Meiji Health Center, Tokyo, Japan) mAbs; and APC-conjugated anti-CD3 (2C11; PharMingen) mAb. Anti-V1 mAb and anti-V5/V1 mAbs were conjugated with FITC using a standard technique. Flow cytometric analysis was performed using a FACScalibur flow cytometer with CellQuest analysis software (Becton Dickinson, Sunnyvale, CA).

Preparation of epidermal sheets and immunohistochemistry

EDTA-separated epidermal sheets were prepared from ears as previously described (34). Briefly, epidermal sheets were fixed in cold acetone for 15 min and then extensively washed with PBS. Two-color immunofluorescence was performed as follows: sheets were initially incubated overnight at 4°C with APC-conjugated anti-CD3 (2C11; PharMingen) and either FITC-conjugated anti-C (GL3; PharMingen), anti-Cß (H57; PharMingen), or anti-V5 (F536; PharMingen) mAb. After thorough rinsing with PBS, the specimens were mounted in PBS/glycerol, coverslipped, and viewed under a fluorescence microscope (Axiovert 100; Zeiss, Oberkochen, Germany) with CellScan image analyzer system (Scanalytic, Billerica, MA). To aquire images of the epidermal sheets stained with FITC- and APC-conjugated mAbs, filters for 546 nm and 470 nm were selected, respectively. The black and white images were incorporated and analyzed. For quantification of DETC densities, the number of DETC was counted in seven fields of 0.256 mm² per specimen; each specimen was prepared from a different distinct mouse, and four to six mice were used for each group. Data were expressed as mean (± SD) density of positive cells per square millimeter.

Analysis of V and V usage and junctional diversities of fetal thymocytes and DETCs

Total RNA was extracted from IECs, fetal thymocytes, or sorted V5⁺ or V5^- cells with TRIzol reagent (Life Technologies) according to the manufacturer’s instructions. The first strand cDNA was synthesized and amplified by RT-PCR. In brief, total RNA was reverse transcribed with 20 pmol of random hexamer primer (Life Technologies) using reverse transcriptase SuperScript II (Life Technologies) according to the manufacturer’s instructions. The cDNA aliquots were amplified by PCR with various combinations of V/C or V/C primers and 2.5 U AmpliTaq Gold DNA polymerase (Perkin-Elmer/Cetus, Norwalk, CT) in a total volume of 50 ml. The reaction buffer consisted of 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, and 0.01% gelatin. The conditions of thermal cycling were as follows: 94°C for 1 min, 54°C for 1 min, and 72°C for 30 s. PCR samples were heated at 94°C for 7 min before the first cycle and the final extension was prolonged to 4 min. PCR primers used in this study were identical with those previously described (35). Southern blot analysis of V/C-PCR products was performed with C2 cDNA probe (MNG8), which binds to all members of Cg genes (data not shown). The C2 probe was labeled with [-³²P]dCTP using the Megaprime DNA Labeling System (Amersham, Arlington Heights, IL) according to the manufacturer’s instructions. Southern blot analysis of V/C-PCR products was performed using as a probe an oligonucleotide common either to J1 (5'-TTCCACAGTCACTTGGGTCCCCA-3') or to J2 (5'-CTCCACAAAGAGCTCTATGCC-3') segments and labeled with [-³²P]ATP using Ready-To-Go T4 polynucleotide kinase (Pharmacia Biotech, Uppsala, Sweeden) according to the manufacturer’s instructions.

To analyze the N region diversities of TCR - and - chains, the RT-PCR products were gel purified using a Qiaex II gel extraction kit (Qiagen, Chatsworth, CA) and subcloned into T vector pCR II (Invitrogen, San Diego, CA). The subclones were sequenced by the dideoxy chain-termination method using a Dye Deoxy Terminator Cycle Sequencing Kit (Amersham) and an Applied Biosystems model 377 DNA sequencer (Foster City, CA).

In vitro culture of DETCs and semiquantitative RT-PCR of cytokine and KGF gene expression

IECs (2.5 x 10⁵ cells; 5–10% DETC) were cultured in 96-well round-bottom plates at 37°C for 40 h in 5 µg/ml Con A. The cultured or freshly isolated IECs were analyzed for cytokine mRNA expression using semiquantitative RT-PCR method. Procedures for total RNA extraction and cDNA synthesis were identical with those described above. PCR conditions were identical with those previously described in (36) but using pairs of primers for hypoxanthine phosphoribosyltransferase (HPRT; control), IL-2, IL-4, IFN-, and KGF. PCR primer sequences were as follows: HPRT sense 5'-GTTTGTTGTTGGATATGCCCTTGAC-3', antisense 5'-GGGGACGCAGCAACTGACATTTCTA-3'; IL-2 sense 5'-TGATGGACCTACAGGAGCTCCTGAG-3', antisense 5'-GAGTCAAATCCAGAACATGCCGCAG-3'; IL-4 sense 5'-CGAAGAACACCACAGAGAGTGAGCT-3', antisense 5'-GCTCATTCATGGTGCAGCTTATCG-3'; IFN- sense 5'-AGCGGCTGACTGAACTCAGATTGTAG-3', antisense 5'-GTCACAGTTTTCAGCTGTATAGGG-3'; KGF sense 5'-CGGAATTCATGCGCAAATGGATACTGACACGG-3', antisense 5'-CGGAATTCTTAGGTTATTGCCATAGGAAG-3'. RT-PCR products were electrophoresed on 1.8% agarose gels and stained with ethidium bromide.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Generation of ES cell clones with a defined deletion at the V1 locus and V1-deficient mice

To avoid possible interference of the PGK-neo cassette at the V1 locus with TCR gene rearrangement and transcription, we used the Cre-loxP system (31, 32) to delete the V1 gene segment. In the construction of a targeting vector, the loxP-flanked PGK-neo gene was replaced with two exons coding the V1 gene (Fig. 1A). The targeting construct was electroporated into E14K ES cells; of the 600 G418-resistant ES colonies screened, 3 PCR-positive ES clones were obtained. To remove the PGK-neo cassette from the targeted allele of the ES clones, pMCcre-puro was electroporated into each ES cell clone; transient expression of Cre recombinase allowed the PGK-neo cassette to be excised from the genome of the targeted ES cells. Homologous recombination and deletion of the PGK-neo cassette was confirmed by Southern blot analysis of PstI-digested genomic DNA from the ES cells using a genomic DNA fragment flanked to the 3' terminus of the targeting construct as a probe (Fig. 1, A and B). Autoradiography showed that the targeted V1 allele and the neo gene-deleted allele were detected as a 1.8-kb band and a 4.3-kb band, respectively, in comparison to a 5.2-kb band from the wild-type allele (Fig. 1B, lanes 1–3). Chimeric mice were generated from the V1-deficient ES cells, and germline-transmitting mice were obtained by repetitive mating of the chimeric mice with C57BL/6 mice. PstI-digested genomic DNA from tails of offspring from the germline-transmitting mice was analyzed by Southern blot analysis (Fig. 1B, lanes 4–6). Wild-type (V1^+/+), heterogyzously mutated (V1^+/-) and homozygously mutated (V1^-/-) mice from littermates generated by mating V1^+/- mice or V1^+/- with V1^-/- mice were used in the following experiments. V1^-/- mice were born at the expected Mendelian ratios and appeared healthy with no apparent anatomical abnormalities.

Markedly impaired development of V5⁺ as well as TCR⁺ fetal thymocytes in V1^-/- mice

Because V1 gene rearrangement and transcription predominates in early fetal thymocyte development (1, 5, 8, 9), we first analyzed the kinetics of fetal thymocyte development in V1^+/+ and V1^-/- mice by flow cytometry. The total number of Thy-1⁺ thymocytes of V1^+/+ and V1^-/- mice were comparable with each other during the fetal to newborn period (GD15 to newborn (GD20); Fig. 2, A and B). However, both the percentages and numbers of TCR⁺ and V5⁺ thymocytes were markedly lower in V1^-/- fetuses during the early fetal period (GD15-GD18), compared with those in V1^+/+ fetuses (Fig. 2, C–F). The numbers of the TCR⁺ thymocytes in V1^-/- mice were nearly 20 times less on GD15 and GD16, and 4 times less on GD18 than those in V1^+/+ mice (Fig. 2G). The numbers of the V5⁺ thymocytes in V1^-/- mice were nearly 60 times less on GD15, 30 times less on GD16, and 10 times less on GD18 than those in V1^+/+ mice (Fig. 2H). But while the number of TCR⁺ thymocytes in newborn V1^-/- mice was indistinguishable from that observed in newborn V1^+/+ mice (Fig. 2C), the numbers of V5⁺ thymocytes remained low in newborn V1^-/- mice compared with newborn V1^+/- mice (Fig. 2D). These results indicate that in the absence of a functional V1 chain, development of control levels of TCR⁺ thymocytes is notably delayed until the late fetal/newborn period; in contrast, the development of the V5⁺ subset of thymocytes is markedly impaired throughout the entire fetal/newborn period.

View larger version (28K): [in this window] [in a new window]   FIGURE 2. Markedly impaired development of V5+ as well as TCR+ fetal thymocytes in V1-/- mice. Fetal thymocytes were obtained from V1+/+ and V1-/- mice at gestation day (GD) 15, 16, 17, and 18, and newborn (NB) mice. Thymocyte suspensions were triple-stained with either FITC-conjugated anti-V5 or FITC-conjugated anti-C in combination with both APC-conjugated anti-CD3 and biotin-conjugated anti-Thy1.2 mAbs, followed by staining them with RED670-conjugated streptavidin, and then flow cytometric analysis. TCR+ (C) and V5+ (D) cells in the Thy-1+ cell population shown. The absolute numbers of Thy-1+ (B), TCR+ (E), and V5+ (F) cells per thymus were calculated from the total number of unstained thymocytes (A), while the percentages of TCR+ (C) and V5+ (D) cells were calculated as proportions of Thy-1+ cells. Relative cell numbers of TCR+ (G) and V5+ (H) cells in a V1-/- fetal thymus were calculated based on the absolute numbers of the corresponding cells in a V1+/- fetal thymus (set as 100). Open symbols and closed symbols represent the data from V1+/+ and V1-/- mice, respectively. Similar results were obtained from three independent experiments and representative data are shown here.

Generation of TCR⁺ DETCs with normal phenotype in the epidermis of V1^-/- mice

To examine the effects of the deletion of the V1 gene on the development of DETCs, we analyzed the density of DETCs in the ear skin from 1-, 8-, and 16-wk-old V1^+/+, V1^+/-, and V1^-/- mice by immunohistochemistry. Epidermal sheets were double-stained with APC-conjugated anti-CD3 mAb in combination with either FITC-conjugated anti-C, anti-Cß, or anti-V5 mAb. DETCs detected in the epidermal sheets from V1^+/+ and V1^+/- mice were comparable to each other in density and phenotype of TCR (data not shown). Interestingly, CD3⁺ and TCR⁺ DETCs with a typical dendritic morphology characteristic of normal DETCs were observed in V1^-/- mice (Fig. 3, E and e). Unexpectedly, despite the profound deficiency in V5⁺ fetal thymocyte precursors of DETCs in V1^-/- mice, the densities of CD3⁺ DETCs in epidermal sheets from 1-wk-old V1^-/- mice were indistinguishable from those in V1^+/- littermates, and by 8 and 16 wk of age, CD3⁺ DETC densities in V1^-/- epidermal sheets were only 14 and 22% lower than in V1^+/- controls (Table I). As previously reported in normal mice (16, 17, 18), virtually all CD3⁺ DETCs from V1^+/- mice were TCR⁺ and V5⁺ (Table I; Fig. 3, A, C, a, and c). In contrast, while all the CD3⁺ DETC from V1^-/- mice were TCR⁺ (Fig. 3, E and e), V5^- DETCs were frequently observed in the CD3⁺ DETCs from V1^-/- mice (Fig. 3, G and g). Interestingly, V1^-/- mice showed an age-associated increase of percent V5⁺ DETCs (Table I); while the density of V5^- DETCs in V1^-/- mice changed minimally between 1 wk and 16 wk (36% increase over 1-wk level at 8 wk, and 10% increase over 1-wk level at 16 wk; Fig. 4), the density of V5⁺ DETCs almost doubled between 1 and 8 wk (191% increase) and almost tripled between 1 and 16 wk (275% increase; Table I, Fig. 4), suggesting that the expansion potential of V5⁺ DETCs in the epidermis is higher than that of V5^- DETCs. Notably, the V5^- DETCs in V1^-/- mice were localized as clusters distinct from the clusters of V5⁺ DETCs. Many of the V5^- DETCs showed a less dendritic shape (Fig. 3, I and i; a cluster located in lower part of Fig. 3, G and g), while most of the V5⁺ DETCs showed a highly dendritic morphology typical of DETCs of normal mice (Fig. 3, H and h; a cluster located in upper part of Fig. 3, G and g). Additionally, some V1⁺ cell clusters were detected within the V5^- DETC clusters in the epidermal sheets from V1^-/- mice (data not shown), whereas no V1⁺ cells could be found in V1^+/- mice (data not shown). In epidermal sheets from either V1^+/- or V1^-/- mice, we could detect no ßTCR⁺ cells (Fig. 3, B, F, b, and f).

View larger version (122K): [in this window] [in a new window]   FIGURE 3. Immunohistochemical analysis of T cells in epidermal sheets from V1+/- and V1-/- mice. Epidermal sheets prepared from 8-wk-old V1+/- mice (A–D, a–d) and V1-/- mice (E–I, e–i) were double-stained with APC-conjugated anti-CD3 (A-I) in combination with FITC-conjugated anti-C (a,e), anti-Cß (b and f), or anti-V5 (c, d, g, h, i) mAb. A V5+ cluster and a V5- cluster were focused on in H and h and I and i, respectively. Original magnification: x200 (A, a, B, b, C, c, E, e, F, f, G, g), x400 (D, d, H, h, I, i). Representative photographs are shown here.

View this table: [in this window] [in a new window]   Table I. Cell density of DETCs in epidermal sheets from V1+/- and V1-/- mice

View larger version (22K): [in this window] [in a new window]   FIGURE 4. Preferential expansion of V5+ DETCs within the epidermis of V1-/- mice. Epidermal sheets from 1-, 8-, and 16-wk-old V1+/- and V1-/- mice were stained with APC-conjugated anti-CD3 and FITC-conjugated anti-V5 mAbs. The densities of CD3+V5+ and CD3+V5- cells were counted and calculated as described in Table I. Data are expressed as mean ± SD (cell density in square millimeters; n = 4–6) .

View larger version (34K): [in this window] [in a new window]   FIGURE 5. Flow cytometric analysis of DETCs from V1+/- and V1-/- mice. A, IECs prepared from 8-wk-old V1+/- and V1-/- mice as described in Materials and Methods were stained with FITC-conjugated anti-V5, PE-conjugated anti-C, and APC-conjugated anti-CD3 mAbs, or with FITC-conjugated anti-Cß, PE-conjugated anti-C and APC-conjugated anti-CD3 mAbs, and analyzed by flow cytometry with gating on CD3+ cells or CD3+TCR+ cells, respectively. Values in the panels indicate percentages of cells within the quadrants and are representative of four separate experiments. B, IECs prepared from 8-wk-old V1+/- and V1-/- mice were stained with FITC-conjugated anti-V1, PE-conjugated anti-C, biotin-conjugated anti-V5, and APC-conjugated anti-CD3 mAbs, followed by staining with RED670-conjugated streptavidin and analysis by flow cytometry with gating set on CD3+TCR+ cells. The values in the panels indicate percentages of cells within each quadrant, and are representative of 3 distinct experiments. C, IECs prepared from 8-wk-old V1+/- and V1-/- mice were stained with either FITC-conjugated anti-17D1 or control mAb in combination with PE-conjugated anti-C and APC-conjugated anti-CD3 mAbs; Staining is shown as histograms gated on CD3+TCR+ cells. Values in the histograms indicate percentages of 17D1+ cells and are representative of results obtained in four separate experiments.

No requirement for the 17D1-defined TCR conformation for DETC development

Monoclonal Ab 17D1 was initially characterized as recognizing TCRs containing canonical V5-J1-C1/V1-D2-J2-C sequences expressed by DETCs but not by any other T cells (30). Recently, Mallick-Wood et al. (29) reported that a TCR composed of V1-J4-C4/V1-D2-J2-C-chains and expressed on DETCs from V5^-/- mice could be also recognized by 17D1. They suggested that such V1/V1-TCRs from V5^-/- mice conserved the same TCR conformation as formed by the canonical V5/V1-TCRs of normal mice. Furthermore, they also raised the possibility that such a specific TCR conformation rather than simple linear epitopes composed of the particular V- and V-chains might be required for optional development and maintenance of DETCs in the epidermis. To test this hypothesis, IECs from V1^-/- mice were stained with 17D1 and analyzed by flow cytometry. As shown in Fig. 5C, virtually no DETCs from V1^-/- mice were 17D1⁺, although 70–80% of TCR⁺ DETCs from V1^+/- mice were 17D1⁺. Thus, the TCR(s) on DETCs from V1^-/- mice did not express an epitope recognized by 17D1, implying that an essential portion of the 17D1 epitope involves the V1-D2-J2 region. Furthermore, these results indicate that the 17D1 epitope is not essential for development and localization of DETC within the skin.

Normal capacities of in vitro cultured DETCs from V1^-/- mice for cytokine and KGF production

It has been demonstrated that the production of KGF and cytokines including IL-2 and IFN- is induced or enhanced in normal DETCs in response to Con A or stressed keratinocytes (19, 25, 26) To examine the capacities of DETCs from V1^-/- mice for cytokine and KGF production, freshly isolated IECs from V1^+/+ and V1^-/- mice were cultured in the presence or absence of the mitogen, Con A. The purity of DETCs in the IECs was 10%; while keratinocytes accounted for the vast majority of the remaining cells, small numbers (<5%) of Langerhans cells were also present. In freshly isolated IECs from both V1^+/+ and V1^-/- mice, IL-2 and IFN- gene expression was weakly detected before culture while KGF and IL-4 gene expression was undetectable (Fig. 6). Con A stimulation enhanced IL-2 gene expression in IECs from both V1^+/+ and V1^-/- mice (Fig. 6). Similarly, IL-2 production enhanced by dish-bound anti-TCR mAb (UC-7) in DETCs from V1^-/- mice was also comparable to that detected in DETCs from V1^+/+ mice, indicating that TCR-mediated signaling is not impaired in DETCs from V1^-/- mice (data not shown). Notably, IL-2, KGF, and IFN- gene expression was induced or enhanced simply by culture in medium alone (i.e., no Con A), implying that DETCs from V1^-/- mice can respond to unknown Ags possibly expressed on stressed keratinocytes during the in vitro culture as previously observed in freshly isolated DETCs and DETC lines from normal mice (16, 19). These results indicate that the DETCs from V1^-/- mice are functionally competent at least in their capacity to up-regulate cytokine and KGF expression in response to mitogen or to other epidermal cells. The cytokine and KGF gene expression were not significantly detected in CD3⁺ cell-depleted IECs from V1^+/+ and V1^-/- mice (data not shown). Thus their gene expression was mostly derived from DETCs.

View larger version (61K): [in this window] [in a new window]   FIGURE 6. Gene expression of cytokines and KGF in freshly isolated and cultured V1+/- and V1-/- DETCs. IECs (2.5 x 105 cells; 5–10% DETCs) from V1+/- or V1-/- mice were cultured in 96-well round-bottom plates at 37°C for 40 h in complete RPMI 1640 medium containing 10% FBS, with or without 5 µg/ml Con A. The cultured or freshly isolated epidermal cells (the majority of which were keratinocytes) were analyzed for cytokine and KGF mRNA expression using semiquantitative RT-PCR as described in the Materials and Methods. The amount of cDNA for each sample was adjusted using a pair of HPRT primers. Similar results were obtained from two independent experiments.

Change of V usage in fetal thymocytes and DETCs from V1^-/- mice

To characterize differences in the T cell repertoires in fetal thymocytes and adult DETCs from V1^-/- and V1^+/+ mice, V/V usage was first analyzed by semiquantitative RT-PCR. No obvious differences in V usage were observed in DETCs from V1^-/- vs V1^+/+ mice (data not shown). As shown in Fig. 7B, V-J2 expression in early fetal thymocyte (GD15–16) from V1^+/+ mice was almost exclusively confined to V1, although V6 was weakly expressed by GD17. Similarly, adult DETC from V1^+/+ mice displayed only V1-J2 expression. As expected, V1-J2 expression was undetectable in both fetal thymocytes and adult DETC from V1^-/- mice; and while no V-J2 expressionwas seen in adult DETC from V1^-/- mice, V6-J2 and V8-J2 were weakly expressed in fetal thymocytes. V-J1 expression is shown in Fig. 7A. In V1^+/+ mice, V1-J1 expression predominated in both fetal thymocytes and adult DETC. However, V6-J1 and V7-J1 expression was apparent in GD15 fetal thymocytes, and this expression, along with that of V2-J1 and V3-J1 was up-regulated atGD16–17. Weak expression of V6-J1 is also seen in adult DETC from V1^+/+ mice. In V1^-/- mice, no V1-J1 expression was seen. Adult DETC from V1^-/- mice expressed predominantly V6-J1, with detectable expression of V7-J1. V6-J1 and V7-J1 expression also predominated in GD15 fetal thymocyte, while at later times, expression of V2, V3, and V8 increase. In summary, the expression of V genes in adult DETC from both V1^+/+ and V1^-/- mice most closely resembles that of GD15 fetal thymocytes, and in the absence of a functional V1 gene, expression of other previously minor V gene products now predominates.

View larger version (44K): [in this window] [in a new window]   FIGURE 7. V/V gene usage in fetal thymocytes and adult DETCs from V1+/+ and V1-/- mice. A and B, Total RNA extracted from fetal thymocytes and IECs of 8-wk-old V1+/+ or V1-/- was reverse transcribed into cDNA and amplified by PCR with primers for C and various V segments. Southern blot analysis of V-C PCR products was performed using an oligonucleotide probe common to the J1 (A) or J2 (B) segment to examine separately the J1- or J2-linked repertoires, respectively. Representative results from two separate experiments are shown. C, Total RNA extracted from sorted V5+ or V5- DETCs from 8-wk-old V1-/- mice was reverse transcribed into cDNA and amplified by PCR with primer sets for various V-C or V-C genes. Southern blot analysis of V-C PCR products was performed using MNG8 as a probe. The Southern blot analysis of the V-C PCR products was conducted using a common oligonucleotide probe for J1 or J2 segments. Similar results were obtained from two independent experiments and the representative data are shown.

DETCs from V1^-/- mice express fetal-type TCR - and -chains with restricted junctional diversities

To examine the diversity of T cell repertoire in DETCs from V1^-/- mice, we sequenced V(D)J junctional regions amplified by PCR of cDNAs prepared from V5⁺ and V5^- DETCs. As shown in Fig. 8A, all in-frame RT-PCR clones amplified at the V5-J1 junction in DETCs from V1^-/- mice showed the canonical sequence observed in normal DETCs (30/30). In-frame RT-PCR clones amplified at the V6-J junction from both V5⁺ (Fig. 8B) and V5^- (Fig. 8C) DETCs predominantly expressed V6-D2-J1 products, and while P nucleotides were present in over 60% of the sequences, N nucleotides were present in <20% of the junction; such junctions are common in fetal thymocytes (1, 37). Notably, an amino acid sequence ((I/V/L)GGIRA) at the V6-D2-J1 junctions was very frequently observed in both V5⁺ and V5^- DETC transcripts (10/17 and 11/20, respectively; data not shown). Such a junctional sequence was previously observed in neonatal thymic hybridomas and DETC lines spontaneously secreting cytokines (38). Only a few V6⁺ clones used D1, D2, and J1 genes, which are typically detected in adult thymocytes (1, 37). Collectively, these data suggest that the diversity of T cell repertoire of DETCs in V1^-/- mice is biased toward relatively simple joints, consistent with the majority of such cells being derived from fetal, rather than adult, sources.

View larger version (45K): [in this window] [in a new window]   FIGURE 8. Junctional sequences of TCR - and -chains expressed in DETCs from V1-/- mice. RT-PCR products prepared for the analysis of V and V usage in IECs from 8-wk-old V1+/+ or V1-/- mice were subcloned into T-vector pCRII (Invitrogen), followed by sequencing the subclones. Nucleotide sequences of V5-J1 junctions in whole DETCs from V1-/- mice (A) and those of V6-J junctions in V5+ (B) and V5- (C) DETCs from V1-/- mice. The sequences are aligned with germline sequences of TCR and segments. The percentage (proportion) of in-frame joints in the V5-J1 (A), V5+ V6-J (B), and V5- V6-J (C) sequences were 94% (30/32), 89% (17/19), and 100% (20/20), respectively. Only sequences with in-frame joints are shown here.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Because the development of V5⁺ fetal thymocytes was markedly impaired in V1^-/- mice, we expected a similarly impaired development of DETCs in these mice. However, TCR⁺ DETCs with a typical dendritic morphology were readily observed in V1^-/- mice, and their density in adult V1^-/- epidermis was only about 20% lower than in V^+/- epidermis (Fig. 3, Table I). Moreover, the TCR⁺ DETCs from V1^-/- mice were capable of up-regulating mRNA for various cytokines and KGF (26) in response to keratinocytes (Fig. 6). These results strongly suggest that the V1 structure is not essential to generate the TCR conformation required for relatively normal DETC development and function. Rather similar results have been shown in V5^-/- mice, where V5^- TCR⁺ DETCs displayed morphologically and functionally normal development (Mallick-Wood et al., Ref. 29). Although these results strongly suggest that neither canonical V5- nor V1-bearing TCRs are essential for DETC development and localization, the possibility remains that TCRs expressing either canonical V5 and/or V1 may constitute more preferential conformations than other TCRs for optimal DETC development/localization, ligand recognition, and/or function. This possibility is consistent with the following observations: 1) canonical V5⁺ DETCs or canonical V1⁺ DETCs are a dominant population in V1^-/- mice or V5^-/- mice, respectively (Table I; Ref. 29); 2) the proportion, and even more strikingly, the absolute numbers of V5⁺ DETC increased in an age-associated manner in V1^-/- mice (Table I); 3) V5^- DETCs displayed less dendritic morphology than V5⁺ DETCs in V1^-/- mice (Fig. 3); 4) the V5⁺ DETCs, but not V5^- DETCs, in V1^-/- mice retained the ability to expand and develop normally in the epidermis comparable to conventional V5⁺ DETCs in normal mice (Fig. 4); and 5) expression levels of TCR on V5^- DETCs in V1^-/- mice are apparently lower than those of V5⁺ DETCs in V1^-/- mice and normal mice (Fig. 5A). The latter three observations are consistent with the possibility that the V5^- DETCs in V1^-/- mice may be less mature than V5⁺ DETCs, as are both our and others’ observations that DETCs in normal mice from birth to 2 wk of age show a round, less dendritic shape and low-level TCR expression compared with adult DETCs (data not shown; Refs. 41 and 42).

The V5⁺ DETCs in V1^-/- mice predominantly expressed TCRs composed of the canonical V5 chain paired with various V6-D2-J1 chains containing relatively few N nucleotide additions (Fig. 8