| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Hans Spemann Laboratories, Max Planck Institute of Immunology, Freiburg, Germany;
Institute of Molecular Biology (Cancer Research), University of Essen Medical School, Essen, Germany; and
Department of Molecular Immunology, Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
strongly reduces CD8 expression. To our knowledge, this is the first demonstrated gene knockdown by morpholino oligonucleotides in primary lymphocytes. Using this approach, we show here that the transcription factor Runx3 is involved in silencing of CD4 expression during CD8 T cell differentiation. Runx3 protein expression appears late in thymocyte differentiation and is confined to mature CD8 single-positive thymocytes, whereas Runx3 mRNA is transcribed in mature CD4 and CD8 thymocytes. Therefore, Runx3 protein expression is regulated at a post-transcriptional level. The knockdown of Runx3 protein expression through morpholino oligonucleotides inhibited the development of CD4-/CD8+ T cells. Instead, mature cells with a CD4+/CD8+ phenotype accumulated. Potential Runx binding sites were identified in the CD4 gene silencer element, which are bound by Runx protein in EMSAs. Mutagenesis of potential Runx binding sites in the CD4 gene silencer abolished silencing activity in a reporter gene assay, indicating that Runx3 is involved in CD4 gene silencing. The experimental approach developed here should be valuable for the functional analysis of other candidate genes in T cell differentiation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain genes, CD4-/CD8- (double-negative (DN)3) T cells develop into CD4+/CD8+ (double positive (DP)) T cells, at which stage rearrangement of the TCR genes takes place. These immature DP cells show a TCRlow phenotype. DP thymocytes then undergo negative or positive selection depending on the capacity of their TCRs to interact with thymic epithelial cells expressing self-peptides bound to MHC molecules. Positively selected thymocytes increase TCR expression and up-regulate the activation marker CD69; they are thus characterized by a CD4+/CD8+/CD69+/TCRint phenotype (3, 4, 5). Further differentiation leads to either mature CD4+/CD8- (single positive (SP)) Th cells activated by MHC class II/Ag complexes or CD4-/CD8+ SP CTLs interacting with MHC class I/Ag complexes. These mature thymocytes have a CD69-/TCRhigh phenotype. The close correlation of CD4 and CD8 gene expression with the differentiation process makes it likely that these mechanisms are functionally connected (6). Both the analysis of genes involved in CD4/CD8 gene regulation and the identification of genes differentially expressed during thymocyte differentiation may provide insight into the mechanisms governing T cell development in the thymus.
Gene function in thymocyte development has mostly been analyzed in time-consuming knockout or transgenic approaches. Functional gene inactivation (knockdown) appears a promising alternative. Here we have developed a novel approach that allows the efficient analysis of gene function during late T cell development. The approach is based on the transfection of ex vivo isolated intermediary DP thymocytes (CD69+/TCRint) with gene-specific antisense morpholino oligonucleotides. Differentiation of the transfected cells is induced by either IL-7 or in reaggregated thymic organ cultures (RTOC). To validate our approach, we show that CD8 expression is strongly reduced through a CD8-specific morpholino antisense oligonucleotide.
Runx transcription factors play important roles in cell differentiation and proliferation. Runx1 (acute myeloid leukemia 1 (AML1)/core binding factor 2 (CBF2)/polyoma enhancer binding protein 2B (PEBP2B)) is a master regulator of hemopoiesis and is frequently mutated in human AML (7). Runx2 (AML3/CBF1/PEBP2A) is essential for bone differentiation (8). Runx3 (AML2/CBF3/PEBP2C) is expressed in the hemopoietic system (9, 10, 11) and appears to be involved in myeloid and B cell differentiation (11, 12). Moreover, Runx3 was recently shown to be a potent suppressor of gastric cancer (13) and to regulate the development of dorsal root ganglia neurons (14). Runx proteins bind to DNA and heterodimerize with the common subunit, CBF (PEBP2) via the highly conserved Runt domain (7, 15). They bind to the consensus core motif PyGPyGGT (7, 15).
Several lines of evidence link Runx proteins to T cell differentiation. Overexpression of Runx1 or Runx2 transgenes in developing thymocytes leads to an increase in immature CD8 SP and DN thymocytes, respectively (16, 17). On the other hand, transgenic expression of a dominant negative form of AML1 (Runx1), which may also inhibit the function of the other Runx proteins, causes a reduction of both CD4 and CD8 SP cells in the thymus (18).
Using morpholino-mediated gene knockdown, we analyze here the role of Runx3 in CD4/CD8 T cell differentiation. We show that Runx3 protein is preferentially expressed in mature CD8 SP thymocytes (CD69-/TCRhigh), but in neither CD4 SP cells nor intermediary DP thymocytes showing signs of positive selection (CD69+/TCRint). Surprisingly, the corresponding Runx3 mRNA is expressed in similar amounts in mature CD4 and CD8 SP thymocytes. Runx3-specific morpholino oligonucleotides specifically block down-regulation of CD4 expression during the development of CD8 SP T cells in an IL-7-dependent cell culture system as well as in reaggregated thymic organ cultures. Instead, an increase in CD4+/CD8+ DP cells with a mature phenotype (TCRhigh/heat-stable Ag (HSA)low/CD69-/CD62Lhigh) and in CD4 SP cells is observed. Stable reporter gene assays with mutated potential Runx binding sites in the CD4 gene silencer revealed that these sites might be involved in CD4 silencing. Taken together these results indicate that Runx3 is involved in the down-regulation of CD4 during thymocyte development.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thymocytes were isolated from 4- to 6-wk-old C57BL/6 mice. Thymocytes were cultured in IMDM with 10% FCS, 150 µg/ml streptomycin, 150 U/ml penicillin, 3 mM L-glutamine and 50 µM ME. TK-1-5 is a subclone of the murine CD4-/CD8+ lymphoma T cell line TK-1, and RLM-11-1 is a subclone of the murine CD4+/CD8- lymphoma T cell line RLM-11 (19, 20). Both lines are CD3+ and TCRhigh. TA3 is a murine B cell hybridoma. TK-1-5 and RLM-11 were cultured in RPMI, and TA3 was cultured in DMEM with 10% FCS, 150 µg/ml streptomycin, 150 U/ml penicillin, and 3 mM L-glutamine.
Thymocyte preparation
Total thymocytes were stained with the following Abs (BD Biosciences, Mountain View, CA): CyChrome-anti-CD4, PE-anti-CD8, FITC-anti-CD69, allophycocyanin-anti-TCR, biotin-anti-HSA, biotin-anti-CD62L, and allophycocyanin-streptavidin. Stained thymocytes were sorted and analyzed by flow cytometry (MoFlo (Cytomation, Fort Collins, CO) and FACSCalibur (BD Biosciences)). For the isolation of CD69+/TCRint R3 cells, CD69+ cells were first enriched from total thymocytes by magnetic sorting (AutoMACS, Miltenyi Biotec, Auburn, CA) and then further sorted by flow cytometry to obtain CD69+/TCRint thymocytes.
IL-7-induced thymocyte development
Sorted CD69+/TCRint (R3) thymocytes (2.5–3 x 106) were electroporated with morpholino oligonucleotides and cultured in complete IMDM. After 3 h, 3 ng/ml IL-7 (rmIL-7; R&D Systems, Minneapolis, MN) was added. After overnight incubation the cells were washed twice with PBS (37°C) and treated for 10 min with 100 µg/ml Pronase (Calbiochem, La Jolla, CA) and 100 µg/ml DNase (Roche, Indianapolis, IN) in 1 ml PBS at 37°C. The reaction was stopped with 100 µl of FCS. The cells were further cultured with IL-7 for 2 days, stained with Abs, and analyzed by flow cytometry. After electroporation, overnight incubation, and subsequent Pronase treatment, including repeated washing and centrifugation steps, 
2–5% of cells were recovered as live cells. During further cultivation in IL-7, the cell numbers stayed relatively constant.
Morpholino antisense oligonucleotides, plasmids, and transfection
The following FITC-coupled morpholino antisense oligonucleotides were used (Gene Tools, Philomath, OR): mo-Runx3-1, 5'-tgctcgggtctacgggaatacgcat; mo-Runx3-2, 5'-gcgggaagcgggaggcgggaaggc; mo-CD8, 5'-gaaagcgggtcaacggtgaggc; and control (-globin), 5'-cctcttacctcagttacaatttata. Both Runx3 oligonucleotides are complementary to sequences in exon 2 of Runx3.
Murine Runx3 cDNA (accession no. NM019732) corresponding to the form transcribed from the proximal promoter (exons 2–6) (21) and a comparable Runx1 construct (accession no. NM009821) were generated by PCR, cloned into cDNA expression vectors, and verified by sequencing.
Sorted CD69+/TCRint (R3) thymocytes were electroporated (Gene Pulser II; Bio-Rad, Hercules, CA) under the following conditions: 3 x 106 cells in 300 µl of RPMI with 25 µM HEPES (pH 7.0), 10–12 nmol of morpholino, 290 V, and 1000 µF (260 V, 950 µF for TK-1-5 and TA3 cells). FITC-positive TK-1-5 cells were sorted after overnight incubation by flow cytometry and electroporated a second time. Protein was isolated after overnight incubation. TA3 cells were cotransfected with Runx1- or Runx3-expressing plasmids and morpholino oligonucleotides and analyzed 24 h after transfection. All transfectants showed similar transfection efficiencies, as judged by the uptake of FITC-labeled morpholino oligonucleotides.
Quantitative real-time PCR
Total RNA of thymocyte populations was prepared with the Absolutely RNA kit (Stratagene, La Jolla, CA). One microgram of total RNA was used to prepare single-strand cDNA with a (dT)15 primer and Expand reverse transcriptase (Roche). The following primers were used for quantitative real-time PCR measuring SYBR Green incorporation (Light-Cycler, Roche): murine Runx3 exons 2–4: forward, 5'-acgctgccggtcgccttca; reverse, 5'-gttcccggggtccatccaca (theoretical product length, 277 bp); murine Runx3 exons 1–2: forward, 5'-ttgtagcagcagggggagaaa; reverse, 5'-agaggaagttggggctgtcg (451 bp); and murine hypoxanthine phosphoribosyltransferase (HPRT): forward, 5'-gctggtgaaaaggacctct; reverse, 5'-cacaggactagaacacctgc (250 bp).
Western blot
Total protein extracts were prepared, and Western blotting was conducted as previously described (22). As primary Abs we used polyclonal rabbit antisera specific for the highly conserved C-terminal 15 aa of murine Runx1 (NMPPARLEEAVWRPY) (23), the 270 aa of the C terminus of human RUNX3 (11) (provided by Y. Groner), the 18 aa of exon 2 of murine Runx3 (Oncogene Research Products, San Diego, CA), or a mouse mAb against the housekeeping genes eIF4 and -actin. As secondary Abs we used peroxidase- or alkaline phosphatase-coupled rabbit- or murine-specific antisera (Amersham Pharmacia Biotech, Santa Cruz, CA). For peptide competition experiments, 20 µl of the Runx1 peptide antiserum (23) was preincubated with 75 µg of the Runx1 peptide for 60 min on ice and then used for Western blotting.
RTOC
RTOC was conducted as previously described (5). Briefly, 14.5-day-old embryonal thymic lobes were cultured for 5 days on a swimming membrane in complete IMDM supplemented with 1.2 mM 2'-deoxyguanosine (Sigma-Aldrich, St. Louis, MO) to kill the thymocytes. Thymic epithelium was disaggregated by trypsin treatment and reaggregated with sorted CD69+/TCRint (R3) thymocytes, which had been transfected with morpholino antisense oligonucleotides several hours previously (3–5 x 104 cells). The aggregates were cultured on a swimming membrane in complete IMDM for 4 days, mechanically disaggregated, stained with Abs, and analyzed by FACS. Some cell loss occurred during the aggregation and desegregation steps, so that 0.5–1 x 104 thymocytes could be recovered from these cultures.
EMSA
Whole cell extracts and EMSA conditions were described previously (24). Extracts were incubated with 4 ng of the double-stranded radioactive Runx-a probe (5'-gccactgaaccacaagggtc) and, where applicable, with a 100-fold molar excess of wild-type or mutated (5'-gccactgaaaaaaaagggtc) unlabeled competitor or with anti-Runx antiserum. The underlined sequences represent wild-type or mutated putative Runx binding sites.
Stable reporter gene assays
Reporter constructs are based on the d2EGFP-N1 plasmid (Clontech, Palo Alto, CA) containing a destabilized version of enhanced green fluorescence protein (EGFP) as a reporter gene. Murine CD4 regulatory elements (25) were generated by PCR from genomic DNA and were verified by sequencing. The 1200-bp chicken -globin insulator element (26) was provided by G. Felsenfeld. Three silencer mutations, MutA, MutB, and MutC, in which guanosines in the TGTGGT sequences of the potential Runx binding sites, Runx-a, -b, or -c, in the CD4 silencer element were replaced by thymidines, were generated by site-directed mutagenesis. Plasmids were linearized upstream of the CD4 enhancer/promoter element before transfection. Each construct was transfected by electroporation into the RLM-11-1 and TK-1-5 T cell lines under the following conditions: 5 x 106 cells in 300 µl of RPMI medium with 25 µM HEPES (pH 7.0), 5 µg of linearized DNA, 75 µg of tRNA, 250 V, and 950 µF. Cells were selected with G418 (TK-1-5, 0.9 mg/ml G418; RLM-11-1, 1.2 mg/ml) for 2–3 wk, and the stably transfected cell pools were analyzed by FACS for d2EGFP expression.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Late T cell differentiation progresses from CD4+/CD8+/CD69-/TCRlow cells to a population of CD69+/TCRint cells. These cells were called R3 by Barthlott and colleagues (5). R3 cells further differentiate via a CD69+/TCRhigh stage (R4) into either mature CD4+/CD8-/CD69-/TCRhigh or CD4-/CD8+/CD69-/TCRhigh SP thymocytes (5) (Fig. 1A). Here, we have adapted the IL-7-dependent in vitro T cell differentiation system of Brugnera and colleagues (27) for the analysis of candidate genes during late T cell differentiation. In contrast to Brugnera and colleagues, we used in vivo positively selected CD69+/TCRint R3 thymocytes as the starting population for IL-7-dependent thymocyte differentiation. After the first night of incubation with IL-7, the cells were subjected to a Pronase treatment to reveal more clearly the dynamic changes in gene expression (27). As shown in Fig. 1A, sorted CD69+/TCRint thymocytes (R3; panels 3 and 4) differentiate preferentially into CD8 SP cells with a mature CD69-/TCRhigh phenotype in the presence of IL-7 (panels 5 and 6). CD69+/TCRint cells therefore appeared to be a good starting population for the establishment of a functional test for CD8 T cell differentiation.
|
(Fig. 1B). FITC-coupled morpholino oligonucleotides permit the identification of transfected and untransfected cells, with the oligonucleotide-negative cells serving as the internal negative control of the experiment. Transfection efficiencies were on the order of 10–30% in different experiments. Sorted CD69+/TCRint cells were transfected with the CD8 morpholino antisense oligonucleotide by electroporation and cultivated with IL-7. Comparison of FITC-negative and FITC-positive cells revealed a clear reduction of CD8 expression in cells transfected with the CD8 morpholino (Fig. 1B). This indicated that the expression of even highly expressed genes such as CD8 can be inhibited in our experimental system. Runx3 protein is expressed in CD8 SP thymocytes, but not in CD4 SP or immature thymocytes
We analyzed Runx3 protein expression in different thymocyte subsets by Western blotting (Fig. 2, A–C). With an antiserum specific for the highly conserved 15 aa in the C terminus of murine Runx1 that also interacts with murine Runx2 and Runx3 (16, 23), we detected a protein of 50 kDa in mature CD8 SP thymocytes, which probably corresponds to Runx3 (Fig. 2A, indicated by the arrow). This protein cannot be detected in mature CD4 SP cells, in intermediary CD4+/CD8+/CD69+/TCRint R3 cells, or in total thymocytes, which consist mostly of CD4+/CD8+/CD69-/TCRlow cells. Preincubation of serum with the peptide used for immunization led to the disappearance of the band (Fig. 2A). Similar results were obtained with an antiserum raised against the C-terminal 270 aa of human RUNX3 that also reacts with murine Runx3 (11) (Fig. 2B). Identification of Runx3 was further confirmed with an antiserum specific for 18 aa encoded by exon 2 of murine Runx3 in the CD4-/CD8+ T cell line TK-1-5. This antiserum detected the same 50-kDa band as the Runx peptide-specific serum used in Fig. 2A (Fig. 2C), but was too weak to detect Runx3 in protein extracts of thymocyte fractions (data not shown). With all three antisera, a single or double band of 56 kDa was recognized, probably corresponding to Runx1 (16) (Fig. 2, A–C). Transfection of Runx1 and Runx3 cDNAs into the B cell line TA3 led to the appearance of a 56- or 50-kDa band, respectively (Fig. 3B). This is further evidence that the Runx protein expressed differentially in thymocytes is indeed Runx3. Runx3 protein expression thus appears late in T cell development and only in CD8 SP T cells.
|
|
All three Runx family members are expressed from two independent promoters, and different splice forms have been described for all three genes (10, 21, 29, 30, 31). We investigated the expression of Runx3 mRNA in various thymocyte subpopulations (Fig. 2, D and E). Runx3 has six exons, with exon 1 being transcribed from the distal promoter 1, and exons 2–6 situated downstream of the proximal promoter 2 (Fig. 2D; nomenclature according to Ref.21). The runt DNA binding domain spans exons 2–4. Runx3 transcripts starting with exon 1 use either a splice acceptor site in exon 2 or,in the case of splice variants, the splice acceptor site of another exon (21). Judging from the calculated molecular masses of the different forms of Runx3, the protein found in CD8 SP cells probably corresponds to a Runx3 form transcribed from promoter 2 containing exons 2–6 or to a form transcribed from promoter 1 containing all six exons (Fig. 2D). Contrary to expectations from the protein expression data, we found by quantitative real-time PCR with a primer pair specific for exons 2 and 4 that this mRNA is present in similar amounts in CD4 SP and CD8 SP thymocytes (Fig. 2E). Gel electrophoresis and sequence analysis revealed a unique amplification product of the expected size and sequence of a Runx3 form containing exons 2, 3, and 4 (data not shown). No transcripts derived from usage of promoter 1 could be detected with a primer pair spanning exons 1 and 2 (data not shown). Therefore, the main Runx3 mRNA in thymocytes is expressed from promoter 2, and Runx3 protein expression must be controlled at the post-transcriptional level.
Morpholino antisense oligonucleotides inhibit Runx3 protein expression
To investigate the role of Runx3 during thymocyte development we attempted to reduce the protein level of Runx3 by morpholino antisense oligonucleotides. We generated a morpholino antisense oligonucleotide that is complementary to a sequence overlapping the AUG in exon 2 of Runx3 (mo-Runx3-1). This oligonucleotide was tested in the murine CD4-/CD8+ T cell line TK-1-5. After transfection of an FITC-coupled control or the Runx3-specific morpholino antisense oligonucleotide into this cell line, the protein level of Runx3 was analyzed by Western blotting with two Runx-specific antisera. Both antisera revealed an almost complete inhibition of Runx3 protein expression by the Runx3-specific morpholino oligonucleotide (Fig. 3A). This oligonucleotide was also tested in TA3 cells transfected with cDNAs coding for murine Runx1or Runx3. The 56-kDa band apparent in Runx1-transfected cells was not affected by either the control or mo-Runx3-1 oligonucleotide, whereas the expression of Runx3 was reduced to background levels by the specific oligonucleotide (Fig. 3B).
Knockdown of Runx3 inhibits down-regulation of CD4 expression during development of CD8 SP thymocytes in an IL-7-dependent cell culture system
The mo-Runx3-1 morpholino antisense oligonucleotide was used to transfect sorted CD69+/TCRint R3 thymocytes to analyze their development in the IL-7 cell culture assay (Fig. 4). Comparison of FITC-negative and FITC-positive cells transfected with a control oligonucleotide specific for -globin showed a similar pattern of CD4 and CD8 expression after IL-7 culture (Fig. 4, panels 2 and 3), with the exception of a slight decrease in CD8 SP cells and a corresponding increase in CD4 SP cells in control morpholino-transfected, FITC-positive cells. Transfection of CD69+/TCRint thymocytes with control morpholino oligonucleotides therefore does not influence T cell differentiation.
|
To test the specificity of our approach, a second Runx3-specific morpholino oligonucleotide (mo-Runx3-2) was tested that is complementary to a region 5' of the AUG in exon 2. This oligonucleotide gave comparable results to the first morpholino oligonucleotide (Fig. 5).
|
Knockdown of Runx3 inhibits down-regulation of CD4 expression during development of CD8 SP thymocytes in RTOC
We also tested the capacity of the morpholino-transfected thymocytes to differentiate in RTOC experiments. Sorted CD69+/TCRint (R3) thymocytes were reaggregated with thymic epithelial cells from 14.5-day-old mouse embryos (5). After 4 days the reaggregated thymi were analyzed by FACS. Most of the CD69+/TCRint thymocytes develop into CD69-/TCRhigh CD4 and CD8 SP T cells (Fig. 6 and data not shown). In this culture system more CD4 SP than CD8 SP thymocytes are normally obtained. The sorted CD69+/TCRint (R3) thymocytes were transfected as before with the mo-Runx3-1 or a control oligonucleotide and then reaggregated with the thymic epithelial cells (Fig. 6, panels 1 and 4). Both the FITC-negative and FITC-positive cells transfected with the control morpholino showed a similar differentiation pattern (Fig. 6, panels 2 and 3). In the cells transfected with mo-Runx3-1, the FITC-negative cells showed a similar differentiation as the control oligonucleotide-transfected cells, whereas the FITC-positive cells showed a severe reduction in CD8 SP cells and an increase in DP and CD4 SP cells (Fig. 6, panels 5 and 6). Repeated experiments led to very similar results (Fig. 6). DP cells in the mo-Runx3-1-transfected cultures showed a mature TCRhigh and CD69- phenotype, indicating again that inhibition of Runx3 function did not interfere with thymocyte maturation, but, rather, counteracted the down-regulation of CD4 during the development of CD8 SP cells (Fig. 6, panels 7 and 8, and data not shown).
|
The silencer element in the first intron of the CD4 gene mediates down-regulation of CD4 expression during the differentiation of CD8 SP thymocytes (25, 34, 35, 36). Analysis of the 434-bp murine CD4 silencer sequence revealed three potential Runx binding sites (TGTGGT), Runx-a, -b, and -c (positions 82–87, 233–238, and 248–253; Fig. 7A). With whole cell extracts of the T lymphoma cell line TK-1-5, a double-stranded probe corresponding to the Runx-a site led to a band shift in EMSAs (Fig. 7B). Binding was efficiently competed for by an unlabeled competitor Runx-a oligonucleotide, but not by a mutated oligonucleotide. Addition of a Runx-specific antiserum induced a supershift indicating that the Runx-a binding site of CD4 silencer is indeed bound by Runx protein in Tk-1-5 (Fig. 7B). The functional importance of these potential Runx binding sites for silencing was analyzed in a reporter gene assay based on the stable transfection of murine T lymphoma cell lines (Fig. 7, C and D). In the mutated constructs (MutA, -B, and -C) the guanosines in the respective potential Runx binding sites were changed by point mutations to thymidines. Stable transfection of the constructs into the murine CD4+CD8- T lymphoma cell line RLM-11-1 showed a uniform expression of the reporter gene regardless of the presence or absence of a wild-type or mutated silencer element in the different constructs (data not shown). The same constructs were used to generate stable transfectants in the murine CD4-/CD8+ T lymphoma cell line TK-1-5. The presence of the CD4 silencer element reduced the expression of the EGFP reporter by 50% (Fig. 7D, lanes 1 and 2). Mutations in the potential Runx binding sites Runx-a and Runx-c (MutA and MutC; lanes 3 and 5) abolished silencing activity completely, while mutation of site Runx-b had only a small effect (lane 4). These results indicate that mutations in potential Runx binding sites in the CD4 gene silencer can reduce the silencing effect.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(Fig. 1B). This is the first demonstration of a morpholino-mediated gene inactivation in primary lymphocytes. Morpholino-dependent fluorescence of transfected R3 cells does not decline significantly during the culture period (data not shown), thus confirming earlier observations that cells differentiating from the DP to the SP stage do not proliferate (38). Recently, gene inactivation in primary T lymphocytes through small interfering RNAs has been reported (39). However, only in vitro stimulated, actively dividing cells could be transfected in that case (39). Our approach therefore presents the advantage that nondividing primary cells can also be targeted efficiently. Using this novel approach, we show here that Runx3 is involved in down-regulation of CD4 expression during CD8 T cell differentiation. Having first shown morpholino-dependent inhibition of Runx3 expression in cell lines (Fig. 3), we observed inhibition of CD4 silencing during CD8 T cell differentiation with two different Runx3-specific morpholino oligonucleotides in an IL-7-dependent T cell differentiation culture system and in RTOC experiments (Figs. 4–6). The fact that two distinct, nonoverlapping oligonucleotides lead to a very similar effect argues for a specific, Runx3-dependent effect. In both types of experiments the newly appearing population of FITC-positive, Runx3 morpholino-transfected CD4+/CD8+ DP cells showed a phenotype indicative of a late maturation stage (TCRhigh, HSAlow, CD69-, CD62Lhigh; Figs. 4–6) (5, 32, 33). This indicates that the Runx3 morpholinos do not simply block the differentiation of R3 cells (TCRint/HSAhigh/CD69+/CD62Llow/-; Figs. 1, 4–6); rather, they appear to interfere with the down-regulation of CD4 expression. The functional maturation status of the Runx3 morpholino-transfected CD4+/CD8+ DP cells could not be tested because of the relatively low cell numbers recovered in this experimental system.
While this manuscript was in preparation, Taniuchi and colleagues (40, 41) reported on the importance of Runx binding sites in the CD4 gene silencer and the function of Runx proteins for CD4 gene repression. Runx3 was found to be important for CD4 gene repression during the generation of CD8 SP CTLs, while Runx1 is required for CD4 gene repression at the CD4-/CD8- DN stage (41). The knockout of Runx2 did not have an effect on thymocyte differentiation (41). These results strongly support the validity of the experimental approach presented here. In agreement with our findings, Taniuchi and colleagues (41) found the expression of Runx3 mRNA to a similar extent in CD4 SP and CD8 SP thymocytes; Runx3 protein expression, however, was not analyzed. We show here that Runx3 protein is expressed differentially during T cell differentiation (Fig. 2). It is found exclusively in mature (CD69-/TCRhigh) CD8 SP thymocytes, but not in late DP thymocytes (CD69+/TCRint R3 cells) or in mature CD4 SP thymocytes. Potentially, Runx3 protein expression in developing thymocytes is therefore not only necessary, but may also be sufficient to induce CD4 gene repression. The discrepancy between mRNA and protein expression also demonstrates that this crucial step of T cell differentiation is regulated by an unknown post-transcriptional mechanism. It has been reported that Runx1 is protected from ubiquitin-mediated degradation through dimerization with the PEBP2 subunit (42). Whether the same is true for Runx3 is not known. However, we found similar expression of PEBP2 mRNA in different thymocyte subpopulations (unpublished observations).
Taniuchi and colleagues (41) found decreased numbers of peripheral Runx3-/- CD8 cells (both CD8 SP and DP) and showed also a reduced proliferation of such cells in response to stimuli. Runx3 may therefore play also a role in survival and/or proliferation of CD8-positive cells. Our observation of relatively increased numbers of CD4 SP cells in Runx3 morpholino-treated cell culture and RTOC experiments may be related to this phenomenon.
Repression of CD4 gene expression is mediated by a silencer element located in the first intron of the CD4 gene (25, 34, 35, 36, 40, 41, 43). Deletion of this element leads to mature CD4+/CD8+ DP cells with cytotoxic potential (35, 36). Several cis-acting elements have been identified within the CD4 gene silencer, and there is a partial redundancy of function of different elements (40, 41, 44, 45, 46, 47).
The 434-bp murine CD4 silencer element contains three potential Runx binding sites (Runx-a, -b, and -c; Fig. 7). Runx-a and Runx-b correspond to sites 2' and 2 reported by Taniuchi and colleagues, respectively (40, 41). Sites 2' and 2 were found to be important for CD4 gene silencing in vivo by knockin point mutations in the CD4 gene locus. Mutation of either site 2' (Runx-a) or 2 (Runx-b) alone had little effect, but the combined mutation of both sites abolished CD4 silencing (41). The Runx-c site was not analyzed in vivo, but was shown to contribute to CD4 silencing in transient reporter assays in the CD4-/CD8+ T cell line 1200M (41). While our results confirm the importance of site Runx-a (site 2') and show also a contribution of site Runx-c to CD4 silencing, mutation of site Runx-b (site 2) did not appear to contribute to silencing in the TK-1-5 cell line (Fig. 7). Taniuchi and colleagues (41) also found differences in the requirements for CD4 silencer elements when comparing the results from the 1200M cell line with those from transgenic experiments. However, taken together, our results substantiate the functional importance of Runx binding sites for CD4 gene silencing.
Runx transcription factors have been shown to be able to activate or repress transcription in a context-dependent manner involving interactions with other regulatory factors (7, 15, 48). Several other cis-acting elements in the CD4 silencer element have recently been shown through homologous recombination to play a role in the CD4 silencing mechanism (40). Therefore, it is probable that other factors binding to the silencer are required in conjunction with Runx proteins for repression. The evidence provided here indicates, however, that Runx3 is to date the only factor acting on the silencer, which is differentially expressed by CD4 and CD8 SP thymocytes. This points to a crucial role of Runx3 expression for the establishment of the CD4 silencing mechanism.
In conclusion, we show here that Runx3 plays an important role in CD4 silencing during CD8 T cell differentiation. The experimental system we have developed here should be useful to study the function of other genes during thymocyte differentiation. In this manner it is, for example, possible to study the function of genes during the late phase of T cell differentiation, even if their deletion by homologous recombination blocks T cell differentiation at an early stage. This approach may be especially powerful in combination with the recently described Notch signaling pathway-dependent cell culture system for T cell differentiation (49). We have recently undertaken a microarray-based gene expression analysis of T cell differentiation and have identified a number of candidate genes for positive selection and lineage decision. These can now be efficiently analyzed through the approach described here.
|
Acknowledgments |
|---|
, and Dr. G. Felsenfeld for the plasmid pJ13-1 containing the insulator element. We thank Dr. T. Boehm for discussion, and Drs. J. Kirberg and K. Eichmann for critical reading of the manuscript. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Marc Ehlers, Institute of Molecular Biology (Cancer Research), University of Essen Medical School, Hufelandstrasse 55, D-45134 Essen, Germany. E-mail: marc.ehlers{at}uni-essen.de. Or to Dr. Viktor Steimle, Département de Biologie, Université de Sherbrooke, 2500 boulevard Université, Sherbrooke, Québec, Canada J1K 2R1. E-mail address: viktor.steimle{at}usherbrooke.ca
3 Abbreviations used in this paper: DN, double negative; aa, amino acid; AML, acute myeloid leukemia; CBF, core binding factor; DP, double positive; EGFP, enhanced green fluorescence protein; HPRT, hypoxanthine phosphoribosyltransferase; HSA, heat-stable Ag; PEBP, polyoma enhancer binding protein; RTOC, reaggregated thymic organ culture; Runx, runt domain factor; SP, single positive.
Received for publication February 21, 2003. Accepted for publication July 23, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-dependent signaling pathway. J. Biol. Chem. 274:21651.3 (AML2) is induced by TGF-1 to bind and activate the mouse germline Ig promoter. J. Immunol. 161:6751.-selection. J. Immunol. 169:2866.4p/47) mediated binding of lymphoma cells to high endothelial venules of mucosa-associated lymph nodes. Eur. J. Cell Biol. 61:290.[Medline]
upstream regulatory sequences. J. Exp. Med. 178:941./CBF-SMMHC chimeric protein is localized both in the cell membrane and nuclear subfractions of leukemic cells carrying chromosomal inversion 16. Leukemia 14:1253.[Medline]
/CBF, is mainly located in the cytoplasm and has an affinity with cytoskeletal structures. Oncogene 15:677.[Medline]
-globin domain serves as an insulator in human erythroid cells and protects against position effects in drosophila. Cell 74:505.[Medline]
B protein interferes with transactivation and blocks granulocytic differentiation of 32Dcl3 myeloid cells. Mol. Cell. Biol. 17:4133.[Abstract]
+CD3+CD4-CD8+ thymocytes in normal mouse thymus. Int. Immunol. 13:313. protects RUNX1/AML1 from ubiquitin-proteasome-mediated degradation. EMBO J. 20:723.[Medline]
)/AML/PEBP2() dependent activation of tissue-specific gene transcription. J. Cell Sci. 113:2221.[Abstract]
-like-1 in vitro. Immunity 17:749.[Medline]This article has been cited by other articles:
|
|
 |
|
  K. F. Wildt, G. Sun, B. Grueter, M. Fischer, M. Zamisch, M. Ehlers, and R. Bosselut The Transcription Factor Zbtb7b Promotes CD4 Expression by Antagonizing Runx-Mediated Activation of the CD4 Silencer J. Immunol., October 1, 2007; 179(7): 4405 - 4414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Clements, S. A. John, and L. A. Garrett-Sinha Impaired Generation of CD8+ Thymocytes in Ets-1-Deficient Mice J. Immunol., July 15, 2006; 177(2): 905 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Yarmus, E. Woolf, Y. Bernstein, O. Fainaru, V. Negreanu, D. Levanon, and Y. Groner Groucho/transducin-like Enhancer-of-split (TLE)-dependent and -independent transcriptional regulation by Runx3 PNAS, May 9, 2006; 103(19): 7384 - 7389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Liu, B. J. Taylor, G. Sun, and R. Bosselut Analyzing Expression of Perforin, Runx3, and Thpok Genes during Positive Selection Reveals Activation of CD8-Differentiation Programs by MHC II-Signaled Thymocytes J. Immunol., October 1, 2005; 175(7): 4465 - 4474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Abe, K. Kohu, H. Ohmori, K. Hayashi, T. Watanabe, K. Hozumi, T. Sato, S. Habu, and M. Satake Reduction of Runx1 Transcription Factor Activity Up-Regulates Fas and Bim Expression and Enhances the Apoptotic Sensitivity of Double Positive Thymocytes J. Immunol., October 1, 2005; 175(7): 4475 - 4482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Grueter, M. Petter, T. Egawa, K. Laule-Kilian, C. J. Aldrian, A. Wuerch, Y. Ludwig, H. Fukuyama, H. Wardemann, R. Waldschuetz, et al. Runx3 Regulates Integrin {alpha}E/CD103 and CD4 Expression during Development of CD4-/CD8+ T Cells J. Immunol., August 1, 2005; 175(3): 1694 - 1705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Fainaru, D. Shseyov, S. Hantisteanu, and Y. Groner Accelerated chemokine receptor 7-mediated dendritic cell migration in Runx3 knockout mice and the spontaneous development of asthma-like disease PNAS, July 26, 2005; 102(30): 10598 - 10603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kohu, T. Sato, S.-i. Ohno, K. Hayashi, R. Uchino, N. Abe, M. Nakazato, N. Yoshida, T. Kikuchi, Y. Iwakura, et al. Overexpression of the Runx3 Transcription Factor Increases the Proportion of Mature Thymocytes of the CD8 Single-Positive Lineage J. Immunol., March 1, 2005; 174(5): 2627 - 2636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Delaire, Y. H. Huang, S. W. Chan, and E. A. Robey Dynamic Repositioning of CD4 and CD8 Genes during T Cell Development J. Exp. Med., December 6, 2004; 200(11): 1427 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Brenner, D. Levanon, V. Negreanu, O. Golubkov, O. Fainaru, E. Woolf, and Y. Groner Loss of Runx3 function in leukocytes is associated with spontaneously developed colitis and gastric mucosal hyperplasia PNAS, November 9, 2004; 101(45): 16016 - 16021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Telfer, E. E. Hedblom, M. K. Anderson, M. N. Laurent, and E. V. Rothenberg Localization of the Domains in Runx Transcription Factors Required for the Repression of CD4 in Thymocytes J. Immunol., April 1, 2004; 172(7): 4359 - 4370. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, The three potential Runx binding sites, Runx-a, -b, and -c. B, Band shift assay with a Runx-a radioactive probe. The indicated cell equivalents of whole cell extracts of the murine CD4-/CD8+ T cell line TK-1-5 were incubated with the probe. Reactions were performed in the absence (lanes 2–4) or the presence of a 100-fold molar excess of unlabeled wild-type (wt; lanes 5–7) or mutated (M; lanes 8–10) competitor Runx-a oligonucleotides (comp.). For supershift experiments the extract was preincubated with an antiserum against the C-terminal, 15-aa peptide of murine Runx1, which also recognizes murine Runx3 (antib.; lane 13). C, Schematic representation of the different reporter constructs. E, CD4 proximal enhancer (343-bp BstXI-AvaII fragment); P, CD4 promoter (492-bp XbaI-XhoI fragment); EGFP, reporter gene d2EGFP; sil, CD4 silencer element (434-bp SacI-XbaI fragment); silM, 434-bp murine CD4 silencer element mutated in the Runx binding site a, b, or c; black vertical bar, chicken