The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Spain, L. M.
Right arrow Articles by Liu, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Spain, L. M.
Right arrow Articles by Liu, P.
The Journal of Immunology, 2002, 168: 127-133.
Copyright © 2002 by The American Association of Immunologists

TCR{beta} Transmembrane Tyrosines Are Required for Pre-TCR Function1

Lisa M. Spain2 and Pinghu Liu

Jerome H. Holland Laboratory for Biomedical Research, American Red Cross, Rockville, MD 20855


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The pre-TCR promotes thymocyte development in the {alpha}{beta} lineage. Productive rearrangement of the TCR{beta} locus triggers the assembly of the pre-TCR, which includes the pT{alpha} chain and CD3 {epsilon}{gamma}{delta}{zeta} subunits. This complex receptor signals the up-regulation of CD4 and CD8 expression, thymocyte proliferation/survival, and the cessation of TCR{beta} rearrangements (allelic exclusion). In this study, we investigate the function of two conserved tyrosine residues located in the TCR{beta} chain transmembrane region of the pre-TCR. We show that replacement of both tyrosines with alanine and expression of the mutant receptor in RAG-1null thymocytes prevents surface expression and abolishes pre-TCR function relative to wild-type receptor. Replacement of both tyrosines with phenylalanines (YF double mutant) generates a complex phenotype in which thymocyte survival and proliferation are severely disrupted, differentiation is moderately disrupted, and allelic exclusion is unaffected. We further show that the YF double mutant receptor is expressed on the cell surface and associates with pT{alpha} and CD3{epsilon} at the same level as does wild-type TCR{beta}, while association of the YF double mutant with CD3{zeta} is slightly reduced relative to wild type. These data demonstrate that pre-TCR signaling pathways leading to proliferation and survival, differentiation, and allelic exclusion are differently sensitive to subtle mutation-induced alterations in pre-TCR structure.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The mature TCR is capable of precisely discriminating between different affinity/avidity interactions with its MHC/peptide ligand, is influenced by costimulatory and inhibitory interactions, and integrates these inputs to induce distinct signaling outcomes including negative selection, lineage choice and differentiation, activation, proliferation, and survival. In contrast, the pre-TCR is envisioned to be an on/off switch, evolved for the detection of productive TCR{beta} rearrangements and the assembly of a potentially useful receptor during thymocyte development. Nevertheless, it is clear that the pre-TCR, like the mature receptor, engages a diverse collection of downstream signaling pathways, and each may mediate distinct or overlapping ultimate events important for thymocyte development (reviewed in Ref. 1).

The structural correlates of function of the pre-TCR are not fully characterized. However, gene disruption studies show that CD3{epsilon} is absolutely essential for pre-TCR function (2, 3), and CD3{gamma} deficiency severely represses differentiation and expansion (4, 5). CD3{zeta}{eta}-deficient thymocytes are variably and incompletely blocked at the CD4-8- (double negative (DN)3) stage (6, 7, 8, 9). CD3{delta} is expressed in thymocytes and is associated with the pre-TCR but is not required for pre-TCR function (10, 11). CD3 subunits contain conserved immunoreceptor tyrosine-containing activation motifs (ITAM), which are phosphorylated by lck/fyn tyrosine kinases and subsequently serve as interaction sites for syk/ZAP-70 kinases bearing SH2 domains. In mature cells, the ITAMs appear to have overlapping functions and contribute to signaling quantitatively (12, 13, 14). Similarly, in the pre-TCR, substitution of normal CD3{zeta}{eta}, CD3{epsilon}, and CD3{gamma} subunits with versions in which the ITAM sequences are deleted or mutated can still assemble and fully promote CD4+8+ (double positive (DP)) differentiation and expansion (13, 15, 16, 17). It appears that the requirement for different subunits may be more dependent on their relative abilities to promote pre-TCR assembly than on any specific signaling function (18). Although the details of the assembly pathway of the pre-TCR are unknown, inferences can be drawn based on the mature TCR. In this case, expression studies using isolated subunit combinations in nonlymphoid cells suggest a model in which CD3{epsilon} pairs with {delta} in a dimer that associates with TCR{alpha} within the cell, while a second CD3{epsilon} pairs with {gamma} to form a dimer that associates with TCR{beta}. TCR{alpha}{beta} heterodimers bridge the CD3{gamma}{epsilon} and CD3{delta}{epsilon} pairs but are not expressed on the cell surface until the CD3{zeta} homodimer joins the complex. The close association of the CD3{epsilon}{gamma} dimer with TCR{beta} is supported by observations that mAbs against CD3{epsilon}, or the CD3{epsilon}{gamma} heterodimer, interfere with the binding of a second mAb specific for the elbow loop region of TCR{beta} (19).

Most of the extracellular domains of TCR{beta} and pT{alpha} are dispensable for function (20). However, engineering a strong endoplasmic reticulum retention signal in the pre-TCR{beta} chain blocked function (21). These experiments suggest that ligand engagement is unnecessary for pre-TCR function. Instead, T cell development may be triggered by pre-TCR assembly and transport from the endoplasmic reticulum/cis-Golgi to plasma membrane compartments, which presumably allows access to essential signaling molecules.

We have previously determined that the transmembrane domain of the TCR{beta} chain, in particular highly conserved tyrosines therein, is critically important for TCR assembly and function in mature T cells. Mutation of both transmembrane tyrosines to phenylalanines allows cell surface expression and maintains subunit composition but disrupts TCR{beta} chain signaling in response to Ag in T hybridomas and mature primary spleen T cells (22, 23). Mutation of either transmembrane Y or both to leucine or alanine disrupts receptor assembly, greatly reducing coimmunoprecipitation of TCR{beta} with CD3{epsilon} and {zeta}, and, correspondingly, cell surface expression and signaling (22). Others have shown that mutation of the C-terminal conserved transmembrane tyrosine to leucine disrupts human TCR assembly with CD3{zeta}, and when Jurkat T thymoma cells are selected that express high levels of the TCR lacking CD3{zeta}, they have reduced apoptotic responses to Ag (24, 25, 26, 27). Although tyrosines can be sites of phosphorylation, the residues under examination in this study are embedded in a lipid bilayer and are unlikely kinase substrates. They are more likely to be sites of protein-protein interactions. Consistent with this possibility, mutation of the relatively polar tyrosine side chains to nonpolar residues, alanine, leucine, or phenylalanine, was disruptive; if no protein contacts were made such mutations would be expected to be more energetically favorable within the bilayer than the native tyrosine residues. In addition, it has been recently shown that polar asparagine residues within an engineered hydrophobic transmembrane domain promote protein homodimerization within the bilayer (28, 29, 30).

Receptor complex assembly, followed by signal transduction involving protein tyrosine kinase and other pathways, are critical features of pre-TCR function (1). There are both similarities and differences between pre-TCR and mature TCR assembly and signaling. If TCR{beta} transmembrane tyrosine mutations affect signaling mechanisms in common between mature and pre-TCR receptors, we would expect they might affect thymocyte development. This prediction is confirmed herein by our findings that TCR{beta} mutations cause defects in pre-TCR function. Mutation of one or both transmembrane tyrosines to leucine or alanine prevents pre-TCR surface expression on Scid.adh cell lines and abrogates pre-TCR function and T cell development. Replacement of both tyrosines with phenylalanines allows surface expression but expresses a complex developmental phenotype in which allelic exclusion is unaffected, CD4+8+ differentiation is reduced, and thymocyte expansion is severely reduced.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice

RAG-1null C57BL/6 inbred mice were obtained from The Jackson Laboratory (Bar Harbor, ME) and bred for timed pregnancies in our facility in sterile housing. Recipient thymic lobes were isolated from C57BL/6 or B10.BR mice obtained from The Jackson Laboratory and bred in our facility. Animal use protocols are reviewed annually by the Institutional Animal Care and Use Committee.

Retroviral constructs

TCR{beta} mutations were made using site-directed mutagenesis as described previously (22, 23). Mutants were shuttled to the MIGR-1 vector (gift of W. Pear, University of Pennsylvania, Philadelphia, PA) and confirmed by sequencing. Retroviral supernatants were generated by transient transfection of Phoenix cells (31) and titered on 3T3 cells before use. Retroviral supernatants were shown to transmit the enhanced green fluorescent protein (GFP) marker to at least 20% of 3T3 cells after exposure to 0.1 ml of retroviral supernatant. Fetal liver infection efficiencies ranged from 5 to 15%.

Cells, Abs, and flow cytometry

Immature thymoma cells derived from SCID mice, Scid.adh, were provided by D. Wiest (Fox Chase Cancer Center, Philadelphia, PA). mAbs used were as follows: biotinylated KJ25 specific for TCR V{beta}3 (BD PharMingen, San Diego, CA), PE-conjugated anti-TCR{beta} H57–59 (Caltag Laboratories, Burlingame, CA), PE-conjugated anti-CD3{epsilon} 2C11 (Caltag Laboratories), allophycocyanin-conjugated anti-CD8 (BD PharMingen), biotinylated anti-CD4 (Caltag Laboratories), streptavidin red 670 (BD Biosciences, Mountain View, CA), streptavidin PerCP (BD PharMingen), PE-conjugated anti-CD25 (BD PharMingen), PE-conjugated anti-CD5 (BD PharMingen), and PE-conjugated anti-V{beta}8 (F23.1; BD PharMingen). Anti-CD3{gamma}{epsilon} (mAb 7D6) was a gift of L. Samelson (National Institutes of Health, Bethesda, MD). CD3{zeta}-specific HMT 3-1, TCR{alpha}-specific H16-C10, and pT{alpha}-specific polyclonal Abs were gifts from D. Wiest). Surface staining was done in 96-well plates using standard procedures. For DNA analysis, cells were surface stained as required, and after final wash they were fixed for 30 min on ice in freshly made 0.5% paraformaldehyde in PBS. Cells were centrifuged and resuspended in 0.1% Triton X-100 in PBS for 3 min, washed two times in PBS, resuspended in 1 µM 4',6'-diamidino-2-phenylindole (DAPI) in PBS, incubated 1 h or overnight in the dark, and analyzed. Analysis was performed on BD Biosciences instruments, including FACScan, FACSCalibur, and LSR, and analyzed using FlowJo software according to standard procedures.

Fetal liver infection and FTOC

Rag-1null mice were set up for breeding and checked for vaginal plugs daily for 4 days. Plugged females were separated (day 0). Fetal livers were harvested on days 14–16 and disrupted, and cells were frozen in FCS/10% DMSO. Cell suspensions were thawed, washed, and plated at 2 x 106 cells/ml with 10% X3 supernatant (source of IL-3; cells were the gift of F. Melchers, Basel Institute for Immunology, Basel, Switzerland) and 1% stem cell factor (PeproTech, Rocky Hill, NJ) in RPMI 1650 complete (10% FCS, gentamicin, glutamine, 5 x 10-5 M 2-ME, 10 mM HEPES, pH 7.2). Recovery after prestimulation (24 h) is generally equal to input. After 24–36 h, cells were washed, counted, and replated in 24-well plates at 106 cells/well, plus 0.5 ml retroviral supernatant, polybrene (6 µg/ml), and 20% X3, 1% stem cell factor. Virus was pelleted onto cells by spinning plates at 700 rpm for 50 min at room temperature. After 24–36 h, cells were harvested, washed, and resuspended at 105–106 cells/35–40 µl. A total of 35 µl of cell suspension was placed into well-spaced wells of a Terasaki dish (Nunc, Rochester, NY). Recipient thymic lobes were irradiated for 2500 rad, and one lobe was added to each well. Plates were inverted and placed into a plastic box with wet paper towels pre-equilibrated in incubator. After 24 h, each lobe was transferred to standard fetal thymic organ culture (FTOC) conditions, fed weekly, and analyzed after 14–20 days.

Immunoprecipitation and recapture assay

Immunoprecipitation and Western blotting were conducted as described elsewhere (23). In summary, Scid.adh cells were isolated by Ficoll (Amersham Biosciences, Piscataway, NJ) gradient centrifugation followed by several washes in PBS. Cells (1 x 107) were lysed in lysis buffer containing digitonin as detergent and then precipitated using protein A-agarose beads prebound to the designated Ab. Protein separation and immunoblotting were conducted using standard techniques using appropriate Abs and visualized with HRP-bound protein A and chemiluminescence.

For recapture assays, Scid.adh cells were washed twice with HBSS and labeled with biotin for 30 min on ice, after which cell viability was consistently >97%. After labeling, the cells were lysed at the density of 5 x 107/ml for 20 min on ice in buffer containing 1% digitonin (high purity; Wako Pure Biochemicals, Osaka, Japan). The lysates were precleared at 4°C for 1 h with protein A-Sepharose beads (Amersham Biosciences). The extracts were then immunoprecipitated for 2 h at 4°C with the anti-TCR{beta} (H57-597) mAb prebound to protein A-Sepharose. The resultant immune complexes were washed three times with 0.2% digitonin washing buffer and once with PBS. The beads were boiled for 5 min in 100 µl of 1% SDS and the SDS-eluted proteins were quenched with 900 µl of 1% Nonidet P-40 lysis buffer. The solution containing the solubilized proteins was reimmunoprecipitated as above with anti-pT{alpha} cytoplasmic tail Ab (gift of D. Wiest). The recaptured immune complexes were resolved on 12% SDS-PAGE, transferred to membrane, and visualized with HRP-conjugated streptavidin (Southern Biotechnology Associates, Birmingham, AL).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
We tested the abilities of a collection of TCR{beta} transmembrane mutants to rescue the early steps of T cell development in RAG-1null thymocytes. We investigated the two conserved tyrosines embedded in the transmembrane domain of TCR{beta}. Tyrosines have large, polar side chains, so to test these features we made replacements with alanine and leucine (small, nonpolar) and phenylalanine (large, nonpolar). RAG-1null fetal liver cells were infected by retroviral vectors encoding the 2B4 TCR{beta} chain wild type (WT), or encoding TCR{beta} with mutations of the conserved tyrosines: Y to A at both positions 265 and 275 (YA double mutant (YA DM)), and Y to F at both positions (YF DM) (Fig. 1Go). We also tested Y to A, L, and F mutations at the C-terminal tyrosine (275). The vector used (MIGR1) contains the enhanced GFP marker expressed from an internal ribosome entry site. Infected fetal liver cells include T cell progenitors and were used to repopulate irradiated thymic lobes in FTOC. After 14–20 days, lobes were harvested and analyzed by flow cytometry.



View larger version (25K):
[in this window]
[in a new window]
  		FIGURE 1. Alignment of TCR{beta} chain transmembrane regions across phylogenetically diverse species. Two tyrosines (marked with arrows) are conserved. Specific double mutants tested in this study are shown.

 
RAG-1null fetal liver cells infected with the empty vector control generated thymocytes blocked at the CD4-8- (DN) stage, with very low thymus cellularity, similar to uninfected controls (Table IGo). The WT TCR{beta}-containing retrovirus increased thymocyte numbers by >10-fold compared with vector-only control (Table IGo). There was an increase in the percentage of GFP+ cells in the WT transductants relative to vector control, indicating a positive selection for GFP- and TCR{beta}-expressing thymocytes during thymus repopulation. GFP+ thymocytes expressing the WT receptor were found to express both CD4 and CD8 consistent with pre-TCR function and further development (Fig. 2Go and Table IGo). CD25 is expressed at high levels in immature thymocytes from RAG-1null mice and also in thymocytes infected by the vector-only retrovirus (Table IGo and Fig. 2Go). WT TCR{beta} expression on RAG-1null thymocytes induced differentiation as shown by a decrease in the percentage of thymocytes expressing CD25 and a decrease in the density of CD25 expression on both DN and DP thymocytes (Table IGo and Fig. 2Go).


View this table:
[in this window]
[in a new window]
  		Table I. Summary1

 


View larger version (47K):
[in this window]
[in a new window]
  		FIGURE 2. CD4/CD8 coreceptor and CD25 expression on RAG-1null thymocytes infected with TCR{beta}-expressing retroviruses. A, CD4 and CD8 expression is displayed for thymocytes gated by forward and side scatter. Representative thymic lobes are shown. The percentage of DP (boxed) is shown for each sample. See Table IGo for average values and statistical analysis. B, CD25 and CD8 expression is displayed for thymocytes gated for live cells and GFP+. Vector control and YA DM thymocytes express high levels of CD25 on CD8- thymocytes. WT thymocytes down-regulate CD25 on CD8- and CD8+ subpopulations, while YF DM thymocytes maintain higher levels of CD25 on each subpopulation compared with WT. See Table IGo for average values and statistical analysis.

 
In contrast, substitution of both tyrosines with alanines (YA DM) failed to rescue thymocyte development (Table IGo and Fig. 2Go). The small gains in thymic cellularity observed for YA DM were not statistically significant, and the percentage of GFP+ thymocytes was similar to vector control, indicating that there was no expansion of YA mutant TCR{beta}-expressing progenitors (Table IGo). CD4 and CD8 expression was not induced (Fig. 2Go), and CD25 expression remained high (Fig. 2Go). Similarly, substitution of a single Y at the C-terminal (275) position with either A or L prevented the ability of the pre-TCR to restore thymus cellularity and development of DP thymocytes (data not shown). Therefore, substitution of a single or both conserved tyrosines with hydrophobic alanine, predicted to be within the membrane-spanning region, was extremely disruptive. Previous studies with mature receptors showed that such mutations reduced or prevented surface expression, disrupted assembly with CD3 subunits, and abrogated signaling (22). Our experiments described below also suggest that Y to A or L mutations prevent pre-TCR surface expression on the surface of immature thymocytes.

Our previous experiments using mature T hybridomas and splenic T cells showed that replacement of the conserved tyrosines with phenylalanine could allow surface expression but disrupt signaling (22, 23). To determine whether YF mutations could similarly affect pre-TCR function, we tested these mutants in thymocytes. Single substitution of Y275 with F was indistinguishable from WT controls (data not shown). However, F substitution of both conserved Ys at 265 and 275 (YF DM) caused a distinct phenotype, as described below.

The YF DM TCR{beta} mutants were less able to restore thymocyte expansion relative to WT TCR{beta}. The average number of thymocytes recovered from YF DM samples was only 32% of WT controls (Table IGo). The percentage of DP thymocytes in YF DM samples was also reduced, although less severely, compared with WT (66% of control; Fig. 2Go and Table IGo). YF DM thymocytes also showed higher levels of CD25 expression on DN and DP thymocytes, relative to WT (Fig. 3Go and Table IGo). In addition, CD5 was expressed at lower levels on YF DM thymocytes compared with WT (Fig. 3Go). Taken together, these changes in expression indicate that there is a defect of maturation in the cells expressing the mutant TCR{beta}. Some of the differences observed between mutant and WT TCR{beta} could be related to the reduction in proliferation. It has been suggested that proliferation is required for the down-modulation of surface molecules such as CD25 (5).



View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 3. CD5 expression is reduced on all subpopulations of YF DM compared with WT transduced thymocytes. Thymocytes were gated for live, GFP+ cells, and subsequently for CD4/8 subpopulations CD8+CD4- (A), CD4+8+ (B), and CD4-8- (C). The expression of CD5 is shown for WT (solid line) and YF DM (dashed line) TCR{beta}-expressing thymocytes.

 
To further explore the underlying basis of reduced thymocyte expansion, we examined proliferation directly in subpopulations of thymocytes using DAPI staining for DNA content. In RAG-1null thymocytes expressing WT receptor, the most actively cycling subpopulation was CD4-CD25+GFP+, in which ~30% of thymocytes were in the S/G2 phases of the cell cycle. In contrast, only ~20% of CD4-CD25+GFP+ thymocytes from YF DM-TCR{beta} or vector control infected thymuses were in S/G2 (Fig. 4GoA). All other populations cycled at the rate of 10% or less, and the percentage of cycling thymocytes did not differ between WT and vector control (data not shown).



View larger version (39K):
[in this window]
[in a new window]
  		FIGURE 4. YF DM show reduced proliferation and increased cell death. A, Reconstituted RAG-1null thymocytes were gated for expression of GFP+ and CD25+, and analyzed for DAPI. Frequency histograms for DAPI stain are shown, and the percentage of cells with >2 N DNA content are indicated. B, Reconstituted RAG-1null thymocytes were analyzed for cell death by 7-AAD staining. Average and SD percentage of 7-AAD staining for four to five lobes is shown.

 
We also determined the fraction of dead cells in reconstituted thymic organ cultures. There was a statistically significant increase in 7-amino actinomycin D (7-AAD)-positive cells in the YF DM samples compared with WT (Fig. 4GoB). Taken together, these experiments provide evidence for a reduction in both proliferation and survival contributing to the reduction in cellularity of YF DM-expressing thymuses.

There was no correlation between the percentage of DP and the number of thymocytes recovered for either WT or mutant samples (Fig. 5Go). Therefore, up-regulation of CD4 and CD8 expression during thymocyte development is not necessarily linked to thymocyte survival/proliferation. The development of DP thymocytes without expansion has been observed previously, in situations where both distal (p53 deficiency (5), Ikaros deficiency (32)), and proximal (Gads deficiency (33)) signaling mediators are disrupted. Therefore, TCR signals controlling DP differentiation vs proliferation/survival are partially distinct and likely bifurcate early in the signaling pathway.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 5. No correlation between percentage of CD4+8+ thymocytes and cellularity of YF DM reconstituted RAG-1null thymocytes. Each symbol represents a single thymic lobe reconstituted by retroviral infection of RAG-1null fetal liver, as indicated. Results are combined from three independent experiments.

 
The pre-TCR induces allelic exclusion at the TCR{beta} locus, preventing the expression of dual TCR specificities. To test whether TCR{beta} mutations could affect this function, we infected normal fetal liver (RAG-1+/+) and reconstituted irradiated FTOCs. To determine whether the YF DM receptor was capable of allelic exclusion, we tested for the expression of endogenous TCR{beta} receptors (the frequently occurring V{beta}8 family) on the surface of GFP+ thymocytes. As expected, we found that V{beta}8 was expressed on vector control infected thymocytes (Fig. 6Go). V{beta}8 receptor expression was suppressed on GFP+ cells of both WT and YF DM infected thymocytes, indicating that the YF DM pre-TCR is effective at mediating allelic exclusion (Fig. 6Go).



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 6. YF DM TCR{beta}-expressing thymocytes exclude the expression of endogenous V{beta}8 receptors. Normal fetal liver cells were transduced with vector alone, WT TCR{beta}, or YF DM TCR{beta} retroviruses, and used to reconstitute irradiated thymus. Live cells were gated for GFP+ and subsequently analyzed for expression of V{beta}8 WT (A), YF DM (B), and vector controls (C).

 
We next used the GFP marker to determine whether pre-TCR function could be correlated with TCR{beta} expression level. Low levels of the TCR{beta} expression on DP thymocytes, together with relatively high background staining of the anti-V{beta}3 reagent on thymocytes from organ culture, precluded the possibility of making direct measurements of TCR{beta} expression. However, the MIGR1 vector expresses both the TCR{beta} insert and the GFP marker protein from a single IRES-containing message driven by the murine stem cell virus retroviral promoter/enhancer. These vectors are susceptible to regulation by the chromosomal site of integration (position effects), leading to interlobe variations in GFP expression levels. We took advantage of these variations to determine whether there was a correlation between high levels of GFP expression—and therefore TCR{beta}—and other developmental parameters.

There was no correlation between GFP means and the number of thymocytes recovered from YF DM infected thymic lobes (Fig. 7GoA). These experiments suggest that quantitative differences in pre-TCR expression, over the range we observed, are unable to overcome the inability of YF DMs to support thymocyte expansion. In contrast, there was a correlation between GFP expression and the formation of DP thymocytes. We divided the WT and YF DM infected populations into three groups based on GFP fluorescence intensity, low (<200), intermediate 200–1000(200–1000), and high (>1000). We determined the GFP mean fluorescence intensity (MFI) and the percentage of DP thymocytes in each subdivided population, and we computed a correlation coefficient for the data (Fig. 7GoB). We found a high degree of correlation (r = 0.56, >1% confidence level) between GFP MFI and the percentage of DP thymocytes recovered for both the WT (data not shown) and YF DM (Fig. 7GoB) infected populations.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 7. Positive correlation between the percentage of DP and the mean of GFP expression, and lack of correlation between thymic cellularity and GFP mean, in YF DM samples. A, Overall GFP mean for each YF DM lobe, considering only lobes in which GFP expression was uniform, was determined and displayed relative to cell number (x10-4). No statistically significant correlation was detected. B, YF DM samples were divided if necessary with gates for low, medium, and high GFP expression. GFP means were calculated for each subpopulation, and the percentage of DP within each subpopulation was determined and displayed. Positive correlation was obtained within the cutoff for significance at 1% confidence. The line was drawn using CricketGraph (Computer Associates, Islandia, NY).

 
We also examined this question conversely. If the failure of YF DMs to make DP thymocytes was in part due to mutation-induced reductions in TCR{beta} expression or a quantitative reduction in the signal induced by mutated receptors, we might expect there to be a selection for higher levels of GFP intensity among YF DM DPs compared with WT DPs. This prediction is confirmed when the data are first gated on the DP subpopulation and the GFP means are calculated. YF DM-bearing DP cells have a MFI of GFP expression significantly higher than that of WT DP cells (GFP MFI 674 for YF DM and 482 for WT; significantly different by t test, p < 0.001).

These data indicate that, with respect to DP development, the YF DM phenotype results from a quantitative reduction in pre-TCR signaling, due to either a reduction in surface expression of TCR{beta} or a reduction in signaling strength. To distinguish between these possibilities, we tested whether the mutants could be expressed on the surface of Scid.adh tumor cells, a cell line derived from SCID mice, which has been previously shown to express pT{alpha} and other CD3 components required for pre-TCR assembly, but shown to lack all endogenous rearranged TCR chains (TCR{beta} and TCR{alpha}) (34). We found that the YF DM TCR{beta} was indeed expressed on the cell surface of Scid.adh cells (Fig. 8GoA). Surface expression of YF DM TCR{beta} was comparable to WT when detected with the pan-{beta} monoclonal H57 (Fig. 8GoA). In addition, surface expression of CD3{epsilon} (2C11) and CD3{gamma}{epsilon} (7D6) was also induced by YF DM as well as or better than WT receptors (Fig. 8GoA). These results indicate that the YF DM TCR{beta} is expressed on the surface, likely in a complex with CD3{epsilon} and {gamma}. We determined the subunit composition of the pre-TCR complex by coimmunoprecipitation (Fig. 8GoB). We found that the CD3{epsilon} and {zeta} chains were coimmunoprecipitable with TCR{beta} from both WT and YF DM, but not from uninfected Scid.adh lines. We noticed a slight reduction (up to 2-fold, relative to WT) in the ability of YF DM TCR{beta} to coimmunoprecipitate with CD3{zeta} in Scid.adh cells (Fig. 8GoB). To determine whether the YF double mutations affected the association of TCR{beta} with pT{alpha}, we performed recapture assays. Scid.adh cells from uninfected, WT, and YF DM clones were surface-biotinylated and immunoprecipitated using anti-TCR{beta} Ab, and the complexes were resolubilized and reprecipitated with antisera against pT{alpha}. Single bands of the appropriate size for pT{alpha}/TCR{beta} dimers were visualized by streptavidin-HRP staining of nondenaturing gel blots. We found that there was no reduction in the amount of pT{alpha} associated with YF DM TCR{beta} compared with WT (Fig. 8GoB).



View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 8. YF DM TCR{beta} is expressed at the cell surface, induces expression of CD3{epsilon} and {zeta}, and coimmunoprecipitates with CD3{epsilon} and CD3{zeta}. A, Scid.adh cells transduced with retroviral vectors expressing WT or YF DM were compared with uninfected controls after staining for CD3{epsilon}, CD3{epsilon}{gamma}, and TCR{beta} as indicated. WT transductants are represented by the heavy line, YF DMs by the thin line, and controls by the dashed line. B, Cells were lysed and normalized based on total protein content, immunoprecipitated using H57 anti-TCR{beta} Ab, separated by reduced PAGE, and immunoblotted as indicated. Controls are a mature T hybridoma, 58{alpha}, with and without expression of WT TCR{beta}, as indicated. pT{alpha} recapture assays were performed as described in Materials and Methods. A single band was observed in the recapture of pT{alpha}/TCR{beta} proteins from biotinylated WT and YF DM cells, and no bands were observed in uninfected control cells, as shown.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have shown that YF DM mutations clearly affect the ability of the pre-TCR to mediate thymocyte developmental progression, proliferation, and survival. Although the pre-TCR is expressed at low levels on the cell surface normally, this low surface expression appears to be a requirement for function. We find no defect in surface expression of the YF DM pre-TCR in Scid.adh cells, suggesting that failure of surface expression alone cannot account for the YF DM defect. Therefore, although we have not yet directly shown any signaling defects in the YF DMs, we think it is likely that the YF DM mutations alter some aspect of pre-TCR signaling function.

CD3 subunits are clearly involved in signaling by the pre-TCR complex through conserved ITAM motifs. In the mature TCR, evidence supports a quantitative role for ITAM function. For example, it was shown in TCR transgenics that TCR specificity (and, presumably, affinity) could affect the ability of CD3{zeta} transgenes which lack one or more ITAM to rescue CD3{zeta} deficiency (35). Furthermore, the propensity for a given TCR to be positively or negatively selected on a given background correlated with the number of ITAMs present, where fewer CD3{zeta} ITAMs appeared able to convert a negatively selecting signal to a positively selecting one. These previous experiments provide a quantitative link among TCR affinity, signaling strength, thymocyte selection, and the number of ITAMs, in the context of the mature TCR. Because the YF DM mutations affect the ability of CD3{zeta} to associate with TCR{beta}, the phenotype we observe could be due to a reduction in signaling strength due to a reduction in the number of engaged ITAMs. Although coimmunoprecipitation experiments show only a very minor deficiency in CD3{zeta} association, it is possible that functional coupling of the receptor to CD3{zeta} is more highly affected. With respect to DP formation, overexpression of the mutant was able to alleviate the defect in function which supports a quantitative model of signaling. Our observations are the first to support the quantitative signaling model with respect to pre-TCR function.

In contrast, the survival and proliferation of thymocytes was not observed to be correlated with pre-TCR expression levels in YF DMs. This suggests that the YF DM pre-TCR may induce signals that differ from WT qualitatively as well as quantitatively. Recent data concerning mature TCR signaling are also inconsistent with a purely quantitative model of signaling (reviewed in Ref. 36). Specifically, it has been shown that mutations in the TCR{alpha} connecting peptide motif prevent the association of CD3{delta} in the mature TCR complex (37). The CD3{delta}-less receptor of connecting peptide motif mutants (or CD3{delta} knockout mice (38)) is expressed on the cell surface and is competent to signal for negative selection but does not mediate positive selection. Additional experiments showed that the requirement for CD3{delta} in positive selection was independent of its ITAM sequence and correlated with the ability to activate extracellular signal-regulated kinase (37, 38). Future experiments will be aimed at discovering whether the YF DM TCR{beta}-containing pre-TCR may also induce qualitatively different signals, which could explain its particular inability to support thymocyte proliferation and survival.

The pre-TCR functions by activation of multiple signaling pathways. For example, activated lck (39, 40) and protein kinase C (PKC) (41) were shown to be equivalent to pre-TCR in rescuing thymocyte development, while activated Ras and Raf restored proliferation, differentiation, and survival, but failed to induce allelic exclusion (42). Further studies showed that differentiation, proliferation, and survival mediated by activated lck required Rho activity, while lck-induced allelic exclusion was Rho independent (43, 44). Defects in thymocyte survival and proliferation by Vav deficiency were completely reversed by activated Rac-1 (45). Taken together, these findings suggest a model for pre-TCR signaling in which allelic exclusion is dependent on pathways downstream of PKC but not RAS, whereas proliferation, survival, and differentiation require PKC, RAS, Rho, and Rac pathways. In the case of the YF DM TCR{beta}, which is more defective for proliferation and survival than for differentiation or allelic exclusion, we speculate that engagement of any or all of the Ras, Rac, and Rho pathways may be reduced, while other PKC-dependent pathways are normally activated. Signaling strength could determine which pathways are activated if the intrinsic thresholds for activation of key components differ. In addition, there could be qualitative differences in YF DM signaling compared with WT, due to subtle structural alterations. Our experiments so far are consistent with both possibilities, but continued evaluation of the structure and function of YF mutants in thymocytes are likely to provide insights into the mechanism of pre-TCR signaling.


   Acknowledgments
 
We thank Guoyan Gao for technical assistance, Wendy Davidson and the members of our laboratory for review of the manuscript, and David Wiest for generosity with reagents and advice.


   Footnotes
 
1 This work was supported by Public Health Service Grant AI36453. Back

2 Address correspondence and reprint requests to Dr. Lisa M. Spain, Jerome H. Holland Laboratory for Biomedical Research, American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855. E-mail address: spainL{at}usa.redcross.org Back

3 Abbreviations used in this paper: DN, double negative; DP, double positive; ITAM, immunoreceptor tyrosine-containing activation motif; FTOC, fetal thymic organ culture; DAPI, 4',6'-diamidino-2-phenylindole; GFP, green fluorescent protein; WT, wild type; DM, double mutant; 7-AAD, 7-amino actinomycin D; MFI, mean fluorescence intensity; PKC, protein kinase C. Back

Received for publication August 17, 2001. Accepted for publication October 25, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kruisbeek, A. M., M. C. Haks, M. Carleton, A. M. Michie, J. C. Zuniga-Pflucker, D. L. Wiest. 2000. Branching out to gain control: how the pre-TCR is linked to multiple functions. Immunol. Today 21:637.[Medline]
  2. DeJarnette, J. B., C. L. Sommers, K. Huang, K. J. Woodside, R. Emmons, K. Katz, E. W. Shores, P. E. Love. 1998. Specific requirement for CD3{epsilon} in T cell development. Proc. Natl. Acad. Sci. USA 95:14909.[Abstract/Free Full Text]
  3. Malissen, M., A. Gillet, L. Ardouin, G. Bouvier, J. Trucy, P. Ferrier, E. Vivier, B. Malissen. 1995. Altered T cell development in mice with a targeted mutation of the CD3-{epsilon} gene. EMBO J. 14:4641.[Medline]
  4. Haks, M. C., P. Krimpenfort, J. Borst, A. M. Kruisbeek. 1998. The CD3{gamma} chain is essential for development of both the TCR{alpha}{beta} and TCR{gamma}{delta} lineages. EMBO J. 17:1871.[Medline]
  5. Haks, M. C., P. Krimpenfort, J. H. van den Brakel, A. M. Kruisbeek. 1999. Pre-TCR signaling and inactivation of p53 induces crucial cell survival pathways in pre-T cells. Immunity 11:91.[Medline]
  6. Love, P. E., E. W. Shores, M. D. Johnson, M. L. Tremblay, E. J. Lee, A. Grinberg, S. P. Huang, A. Singer, H. Westphal. 1993. T cell development in mice that lack the {zeta} chain of the T cell receptor. Science 261:918.[Abstract/Free Full Text]
  7. Ohno, H., T. Aoe, S. Taki, D. Kitamura, Y. Ishida, K. Rajewsky, T. Saito. 1993. Developmental and functional impairment of T cells in mice lacking CD3{zeta} chains. EMBO J. 12:4357.[Medline]
  8. 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-{zeta}-/- mutant mice and identification of a novel T cell population in the intestine. EMBO J. 12:4863.[Medline]
  9. 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 the CD3-{zeta}/{eta} gene. EMBO J. 12:4347.[Medline]
  10. Dave, V. P., Z. Cao, C. Browne, B. Alarcon, G. Fernandez-Miguel, J. Lafaille, A. de la Hera, S. Tonegawa, D. J. Kappes. 1997. CD3{delta} deficiency arrests development of the {alpha}{beta} but not the {gamma}{delta} T cell lineage. EMBO J. 16:1360.[Medline]
  11. Berger, M. A., V. Dave, M. R. Rhodes, G. C. Bosma, M. J. Bosma, D. J. Kappes, D. L. Wiest. 1997. Subunit composition of pre-T cell receptor complexes expressed by primary thymocytes: CD3{delta} is physically associated but not functionally required. J. Exp. Med. 186:1461.[Abstract/Free Full Text]
  12. Irving, B. A., A. C. Chan, A. Weiss. 1993. Functional characterization of a signal transducing motif present in the T cell antigen receptor {zeta} chain. J. Exp. Med. 177:1093.[Abstract/Free Full Text]
  13. Ardouin, L., C. Boyer, A. Gillet, J. Trucy, A. M. Bernard, J. Nunes, J. Delon, A. Trautmann, H. T. He, B. Malissen, M. Malissen. 1999. Crippling of CD3-{zeta} ITAMs does not impair T cell receptor signaling. Immunity 10:409.[Medline]
  14. van Oers, N. S., P. E. Love, E. W. Shores, A. Weiss. 1998. Regulation of TCR signal transduction in murine thymocytes by multiple TCR {zeta}-chain signaling motifs. J. Immunol. 160:163.[Abstract/Free Full Text]
  15. Shores, E. W., T. Tran, A. Grinberg, C. L. Sommers, H. Shen, P. E. Love. 1997. Role of the multiple T cell receptor (TCR)-{zeta} chain signaling motifs in selection of the T cell repertoire. J. Exp. Med. 185:893.[Abstract/Free Full Text]
  16. Haks, M. C., T. A. Cordaro, J. H. van den Brakel, J. B. Haanen, E. F. de Vries, J. Borst, P. Krimpenfort, A. M. Kruisbeek. 2001. A redundant role of the CD3{gamma}-immunoreceptor tyrosine-based activation motif in mature T cell function. J. Immunol. 166:2576.[Abstract/Free Full Text]
  17. Sommers, C. L., J. B. Dejarnette, K. Huang, J. Lee, D. El-Khoury, E. W. Shores, P. E. Love. 2000. Function of CD3{epsilon}-mediated signals in T cell development. J. Exp. Med. 192:913.[Abstract/Free Full Text]
  18. Malissen, B., L. Ardouin, S. Y. Lin, A. Gillet, M. Malissen. 1999. Function of the CD3 subunits of the pre-TCR and TCR complexes during T cell development. Adv. Immunol. 72:103.[Medline]
  19. Ghendler, Y., A. Smolyar, H. C. Chang, E. L. Reinherz. 1998. One of the CD3{epsilon} subunits within a T cell receptor complex lies in close proximity to the C{beta} FG loop. J. Exp. Med. 187:1529.[Abstract/Free Full Text]
  20. Irving, B. A., F. W. Alt, N. Killeen. 1998. Thymocyte development in the absence of pre-T cell receptor extracellular immunoglobulin domains. Science 280:905.[Abstract/Free Full Text]
  21. O’Shea, C. C., A. P. Thornell, I. R. Rosewell, B. Hayes, M. J. Owen. 1997. Exit of the pre-TCR from the ER/cis-Golgi is necessary for signaling differentiation, proliferation, and allelic exclusion in immature thymocytes. Immunity 7:591.[Medline]
  22. Kunjibettu, S., S. Fuller-Espie, G. B. Carey, L. M. Spain. 2001. Conserved transmembrane tyrosine residues of the TCR {beta} chain are required for TCR expression and function in primary T cells and hybridomas. Int. Immunol. 13:211.[Abstract/Free Full Text]
  23. Fuller-Espie, S., P. Hoffman Towler, D. L. Wiest, I. Tietjen, L. M. Spain. 1998. Transmembrane polar residues of TCR {beta} chain are required for signal transduction. Int. Immunol. 10:923.[Abstract/Free Full Text]
  24. Rodriguez-Tarduchy, G., A. G. Sahuquillo, B. Alarcon, R. Bragado. 1996. Apoptosis but not other activation events is inhibited by a mutation in the transmembrane domain of T cell receptor {beta} that impairs CD3{zeta} association. J. Biol. Chem. 271:30417.[Abstract/Free Full Text]
  25. Teixeiro, E., A. Garcia-Sahuquillo, B. Alarcon, R. Bragado. 1999. Apoptosis-resistant T cells have a deficiency in NF-{kappa}B-mediated induction of Fas ligand transcription. Eur. J. Immunol. 29:745.[Medline]
  26. Sahuquillo, A. G., A. Roumier, E. Teixeiro, R. Bragado, B. Alarcon. 1998. T cell receptor (TCR) engagement in apoptosis-defective, but interleukin 2 (IL-2)-producing, T cells results in impaired ZAP70/CD3-{zeta} association. J. Exp. Med. 187:1179.[Abstract/Free Full Text]
  27. San Jose, E., A. G. Sahuquillo, R. Bragado, B. Alarcon. 1998. Assembly of the TCR/CD3 complex: CD3 {epsilon}/{delta} and CD3 {epsilon}/{gamma} dimers associate indistinctly with both TCR {alpha} and TCR {beta} chains: evidence for a double TCR heterodimer model. Eur. J. Immunol. 28:12.[Medline]
  28. Bowie, J. U.. 2000. Understanding membrane protein structure by design. Nat. Struct. Biol. 7:91.[Medline]
  29. Choma, C., H. Gratkowski, J. D. Lear, W. F. DeGrado. 2000. Asparagine-mediated self-association of a model transmembrane helix. Nat. Struct. Biol. 7:161.[Medline]
  30. Zhou, F. X., M. J. Cocco, W. P. Russ, A. T. Brunger, D. M. Engelman. 2000. Interhelical hydrogen bonding drives strong interactions in membrane proteins. Nat. Struct. Biol. 7:154.[Medline]
  31. Lorens, J. B., M. K. Bennett, D. M. Pearsall, W. R. Throndset, A. B. Rossi, R. J. Armstrong, B. P. Fox, E. H. Chan, Y. Luo, E. Masuda, et al 2000. Retroviral delivery of peptide modulators of cellular functions. Mol. Ther. 1:438.[Medline]
  32. Winandy, S., L. Wu, J. H. Wang, K. Georgopoulos. 1999. Pre-T cell receptor (TCR) and TCR-controlled checkpoints in T cell differentiation are set by Ikaros. J. Exp. Med. 190:1039.[Abstract/Free Full Text]
  33. Yoder, J., C. Pham, Y. M. Iizuka, O. Kanagawa, S. K. Liu, J. McGlade, A. M. Cheng. 2001. Requirement for the SLP-76 adaptor GADS in T cell development. Science 291:1987.[Abstract/Free Full Text]
  34. Carleton, M., N. R. Ruetsch, M. A. Berger, M. Rhodes, S. Kaptik, D. L. Wiest. 1999. Signals transduced by CD3{epsilon}, but not by surface pre-TCR complexes, are able to induce maturation of an early thymic lymphoma in vitro. J. Immunol. 163:2576.[Abstract/Free Full Text]
  35. Love, P. E., J. Lee, E. W. Shores. 2000. Critical relationship between TCR signaling potential and TCR affinity during thymocyte selection. J. Immunol. 165:3080.[Abstract/Free Full Text]
  36. Hogquist, K. A.. 2001. Signal strength in thymic selection and lineage commitment. Curr. Opin. Immunol. 13:225.[Medline]
  37. Werlen, G., B. Hausmann, E. Palmer. 2000. A motif in the {alpha}{beta} T-cell receptor controls positive selection by modulating ERK activity. Nature 406:422.[Medline]
  38. Delgado, P., E. Fernandez, V. Dave, D. Kappes, B. Alarcon. 2000. CD3{delta} couples T-cell receptor signalling to ERK activation and thymocyte positive selection. Nature 406:426.[Medline]
  39. Sohn, S. J., K. A. Forbush, X. C. Pan, R. M. Perlmutter. 2001. Activated p56lck directs maturation of both CD4 and CD8 single-positive thymocytes. J. Immunol. 166:2209.[Abstract/Free Full Text]
  40. Mombaerts, P., S. J. Anderson, R. M. Perlmutter, T. W. Mak, S. Tonegawa. 1995. An activated lck transgene promotes thymocyte development in RAG-1 mutant mice. Immunity 1:261.
  41. Michie, A. M., J. W. Soh, R. G. Hawley, I. B. Weinstein, J. C. Zuniga-Pflucker. 2001. Allelic exclusion and differentiation by protein kinase C-mediated signals in immature thymocytes. Proc. Natl. Acad. Sci. USA 98:609.[Abstract/Free Full Text]
  42. Gartner, F., F. W. Alt, R. Monroe, M. Chu, B. P. Sleckman, L. Davidson, W. Swat. 1999. Immature thymocytes employ distinct signaling pathways for allelic exclusion versus differentiation and expansion. Immunity 10:537.[Medline]
  43. Henning, S. W., R. Galandrini, A. Hall, D. A. Cantrell. 1997. The GTPase Rho has a critical regulatory role in thymus development. EMBO J. 16:2397.[Medline]
  44. Galandrini, R., S. W. Henning, D. A. Cantrell. 1997. Different functions of the GTPase Rho in prothymocytes and late pre-T cells. Immunity 7:163.[Medline]
  45. Gomez, M., V. Tybulewicz, D. A. Cantrell. 2000. Control of pre-T cell proliferation and differentiation by the GTPase Rac-I. Nat. Immunol. 1:348.[Medline]



This article has been cited by other articles:


Home page
 BloodHome page
H. Xu, M. S. Williams, and L. M. Spain
Patterns of expression, membrane localization, and effects of ectopic expression suggest a function for MS4a4B, a CD20 homolog in Th1 T cells
Blood, March 15, 2006; 107(6): 2400 - 2408.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
B. D. Wines, H. M. Trist, R. C. Monteiro, C. van Kooten, and P. M. Hogarth
Fc Receptor {gamma} Chain Residues at the Interface of the Cytoplasmic and Transmembrane Domains Affect Association with Fc{alpha}RI, Surface Expression, and Function
J. Biol. Chem., June 18, 2004; 279(25): 26339 - 26345.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
C. Couedel, E. Lippert, K. Bernardeau, M. Bonneville, and F. Davodeau
Allelic Exclusion at the TCR{delta} Locus and Commitment to {gamma}{delta} Lineage: Different Modalities Apply to Distinct Human {gamma}{delta} Subsets
J. Immunol., May 1, 2004; 172(9): 5544 - 5552.
[Abstract] [Full Text] [PDF]

Home page
 JEMHome page
T. Sasada, M. Touma, H.-C. Chang, L. K. Clayton, J.-h. Wang, and E. L. Reinherz
Involvement of the TCR C{beta} FG Loop in Thymic Selection and T Cell Function
J. Exp. Med., June 3, 2002; 195(11): 1419 - 1431.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Spain, L. M.
Right arrow Articles by Liu, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Spain, L. M.
Right arrow Articles by Liu, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE <