CD4 and CD8 Single-Positive Thymocytes Directs Maturation Of

Home , Cytotoxic T cell

Activated p56lck Directs Maturation of Both CD4 and CD8 Single-Positive Thymocytes Sue J. Sohn, Katherine A. Forbush, Xiao Cun Pan and Roger M. Perlmutter This information is current as of October 1, 2021. J Immunol 2001; 166:2209-2217; ; doi: 10.4049/jimmunol.166.4.2209 http://www.jimmunol.org/content/166/4/2209 Downloaded from References This article cites 57 articles, 18 of which you can access for free at: http://www.jimmunol.org/content/166/4/2209.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021 Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Activated p56lck Directs Maturation of Both CD4 and CD8 Single-Positive Thymocytes1

Sue J. Sohn,2* Katherine A. Forbush,* Xiao Cun Pan,* and Roger M. Perlmutter† p56lck is a protein tyrosine kinase expressed throughout T cell development. It associates noncovalently with the cytoplasmic domains of the CD4 and CD8 coreceptor molecules and has been implicated in TCR signaling in mature T cells. Its role in early thymocyte differentiation has been demonstrated in vivo, both by targeted gene disruption and by transgene expression. Previ- ously, we showed that expression of a dominant-negative form of p56lck in double-positive thymocytes inhibits positive selection. We now demonstrate that expression of constitutively activated p56lck (p56lckF505) accelerates the transition from the double- positive to the single-positive stage. Importantly, p56lckF505 drives survival and lineage commitment of thymocytes in the absence of TCR engagement by appropriate MHC molecules. These results indicate that activation of p56lck constitutes an early step in conveying maturational signals after TCR ligation by a positively selecting ligand. Our study provides direct in vivo evidence for the role of p56lck in regulating TCR signaling. The Journal of Immunology, 2001, 166: 2209–2217. Downloaded from

aturation of thymocytes from CD4ϩCD8ϩ double- in Ref. 15). For example, constitutive expression of the CD8 mol- positive (DP)3 to CD4ϩ or CD8ϩ single-positive (SP) ecule permits maturation of CD4ϩ helper T cells (CD4ϩCD8Tgϩ) stage requires signals generated by the concerted in- bearing an MHC class I-restricted TCR (16, 17). Conversely, con- M ϩ teraction of surface TCRs and CD4 or CD8 coreceptors and the stitutive expression of CD4 molecules rescues CD8 cytotoxic T appropriate MHC molecule (reviewed in Refs. 1 and 2). DP thy- cells bearing a class II-specific TCR (18). Thus, thymocytes ex- http://www.jimmunol.org/ mocytes that interact with self-MHC with sufficient affinity survive pressing class I-restricted TCRs were shown to be positively se- and proceed through final maturation steps (positive selection). T lected in the absence of CD8 when cultured with peptide variants cells that fail to interact or that bind MHC molecules too strongly that artificially improved TCR-MHC interactions (19, 20). These undergo apoptosis (death by neglect and negative selection, re- studies indicate that manipulations that change interactions among spectively) (reviewed in Ref. 2). The transition from the DP to SP the TCR, coreceptor, and MHC invoke differential intracellular ϩ stage is also accompanied by commitment to either the CD4 signals to direct T cell maturation. ϩ helper or CD8 cytotoxic T cell lineage, a decision influenced by Following the TCR ␣ (␣) chain gene rearrangement, the DP the Ag specificity of the TCR and the appropriate MHC-coreceptor thymocytes undergoing selection express clonotypic ␣- and the match. For example, transgenic mice constitutively expressing ␤-polypeptide chains which comprise a structure that can bind the by guest on October 1, 2021 MHC class I-restricted TCRs or those expressing MHC class II- peptide/MHC complex. This heterodimeric receptor also associ- ϩ ϩ restricted TCRs possess mainly CD8 or CD4 T cells, respec- ates noncovalently with the invariant CD3 subunits and the ␨ tively (3–6). Conversely, mice deficient for MHC class I or class chains, which are required for transducing intracellular signals (21, ϩ ϩ II expression lack CD8 or CD4 T cells (7–10). Finally, mice 22). CD4 and CD8 coreceptors stabilize TCR-MHC interactions lacking CD4 or CD8 coreceptors possess greatly reduced numbers and enhance intracellular signaling events triggered by the TCR. of helper or cytotoxic lineage T cells (11–14). Thus, co-cross-linking CD4 and CD3 has been shown to enhance Although TCRs, coreceptors, and MHC molecules clearly direct the ability of T cells to flux Ca2ϩ relative to CD3 cross-linking the transition from the DP to SP stage, it is unlikely that a “correct” alone (23). In contrast, independent cross-linking of CD4 and CD3 combination of these components sends a strictly instructive signal decreased the Ca2ϩ response. These results indicate that corecep- which simultaneously dictates survival and lineage commitment. tors can contribute to the overall amplitude and/or quality of sig- Several studies support the notion that lineage commitment occurs nals transmitted by the TCR. Furthermore, expression of a chi- first and randomly but that survival depends on sustained, optimal meric coreceptor molecule containing the extracellular and coreceptor/TCR interactions with the MHC molecules (reviewed transmembrane domains of CD8␣ and the cytoplasmic domain of the CD4 molecule in F5 TCR (specific for a viral nucleoprotein b *Department of Immunology, University of Washington, Seattle, WA 98195; and peptide presented by H-2D ) transgenic mice promoted the matu- †Merck Research Laboratories, Rahway, NJ 07065 ration of CD4ϩ T cells expressing this class I-restricted TCR (24). Received for publication August 7, 2000. Accepted for publication November These results suggest that the cytoplasmic tail of the CD4 core- 15, 2000. ceptor can influence the outcome of the thymocyte cell fate The costs of publication of this article were defrayed in part by the payment of page decision. charges. This article must therefore be hereby marked advertisement in accordance lck with 18 U.S.C. Section 1734 solely to indicate this fact. The src family protein tyrosine kinase p56 is expressed 1 This work was supported in part by grants from the National Institutes of Health. mainly in T-lineage cells throughout T cell development (review in R.M.P. was an Investigator of the Howard Hughes Medical Institute. Ref. 25). It associates noncovalently with the cytoplasmic tails of 2 Address correspondence and reprint requests to Dr. Sue J. Sohn, Department of the CD4 and CD8 coreceptor molecules and becomes catalytically Molecular and Cell Biology, 465 LSA, University of California, Berkeley, CA 94720. activated when the coreceptors are cross-linked (26–30). Bio- E-mail address: [email protected] chemical evidence suggests that p56lck can also enhance signals 3 Abbreviations used in this paper: DP, double-positive; SP, single-positive; DN, double- ␤ ␤ mediated through the TCR/CD3 complex (31–33), although a di- negative; 2m, 2-microglobulin; HSA, heat-stable Ag; dLGF, distal promoter driving the lck gene bearing the F505 mutation; RAG, recombinase-activating gene. rect physical association between p56lck and TCR/CD3 complex

has not been established. More importantly, lack of functional AG-3Ј and the reverse primer sequence is 5Ј-CAG AGG GCA GAG GTG p56lck abrogated mature T cell activation in a mutant Jurkat sub- AGA CAG-3Ј. The absence of amplification product was scored as ho- clone (JCaM1), providing genetic evidence for the role of p56lck in mozygous disruption in each case. TCR signaling (34). In mice, targeted disruption of the lck gene, or Analyses of transgene expression by immunoblotting transgenic expression of catalytically inactive p56lck under the Transgene expression was determined by immunoblotting total lysates from control of the lck proximal promoter (which drives transgene ex- 5 Ϫ Ϫ thymocytes and splenocytes. Briefly, 5 ϫ 10 cells were lysed in buffer con- pression in the CD4 CD8 double-negative (DN) and DP thymo- taining 1% Triton X-100 (50 mM Tris (pH 7.5), 50 mM NaCl, 1 mM PMSF, cytes), interferes with both cellular expansion and allelic exclusion ␮ ␮ 1 g/ml aprotinin, 1 g/ml leupeptin, 1 mM Na3VO4), and the resulting ly- at the TCR ␤-chain gene locus during the transition from the DN sates were resolved in 10% SDS-PAGE, transferred onto nitrocellulose, blot- to the DP stage (35, 36). In addition, expression of an activated ted with affinity-purified anti-p56lck monoclonal reagent (3A5; Santa Cruz Bio- ␮ form of p56lck (p56lckF505) under the control of the lck proximal technology, Santa Cruz, CA) at 0.25 g/ml followed by incubation with HRP- conjugated sheep anti-mouse secondary Ab (Amersham Life Sciences, promoter arrests thymocyte differentiation before the DP stage and Arlington Heights, IL) at 1/3000 dilution, and visualized by chemilumines- at high levels causes transformation and accumulation of immature cence. Relative intensity of the signals was determined by densitometric anal- thymocytes (37, 38). yses using Molecular Imager system (Bio-Rad Laboratories, Hercules, CA). Data obtained from these studies document the importance of p56lck in delivering signals from the pre-TCR complex, but they do Flow cytometry not provide direct evidence for the role of p56lck in later stages of Thymocytes and splenocytes were stained and analyzed by flow cytometry T cell development. To study the role of p56lck in the maturation according to standard protocols. After RBC lysis treatment, 106 cells were of SP thymocytes, we have previously used the lck distal promoter, incubated in staining media (HBSS-BSA at 10 mg/ml) containing appro-

priate concentrations of fluorochrome-conjugated mAbs on ice, washed, Downloaded from which is active mainly in mature thymocytes and peripheral T incubated with secondary reagent where required, and fixed in 1% para- lck cells, to drive the expression of a dominant-negative form of p56 formaldehyde (PBS, pH 7.4). Flow cytometry was performed on FACScan in mice (39). Using the lck distal promoter circumvents the early (Becton Dickinson, San Jose, CA) and analyzed using ReproMac software developmental block that occurs after expression of a dominant- (TrueFacts Software, Seattle, WA). The following monoclonal reagents ⑀ negative lck transgene under the control of the proximal promoter. were purchased from PharMingen (San Diego, CA): biotin-anti-CD3 (145-2C11); biotin-anti-CD69 (H1.2F3); biotin-anti-CD4 (H129.19); bi- These mice possess reduced numbers of SP thymocytes, indicating otin-anti-heat-stable Ag (HSA); PE-anti CD8␣ (53-6.7); FITC-anti ␤ TCR that dominant-negative p56lck inhibits positive selection. However, (H57-597), FITC-anti-Qa-2 (1-1-2). PE-anti-CD4 (CT-CD4), FITC-anti- http://www.jimmunol.org/ these results do not resolve whether activation of p56lck is suffi- CD8␣ (CT-CD8a), and Tri-Color-conjugated streptavidin were purchased cient to direct survival and/or lineage commitment during the tran- from Caltag Laboratories (Burlingame, CA). For intracellular staining of TCR ␤, cells were first stained for surface sition from the DP to SP stage. In the present study, we used the markers using biotinylated and PE-conjugated monoclonal reagents, re- lck distal promoter to drive expression of an activated form of spectively, followed by incubation with Tri-Color-streptavidin, fixed in 4% p56lck (p56lckF505) in mice. Our data suggest that activation of paraformaldehyde (PBS, pH 7.4), and permeabilized in 0.1% saponin so- lck p56 can direct survival and lineage commitment during the final lution (PBS (pH 7.4)-1% FCS-0.1% NaN3). The cells were then incubated ␤ stages of thymocyte maturation. Furthermore, we show that CD4: with FITC-anti- TCR diluted in permeabilization buffer and analyzed on FACScan. CD8 ratios of developing thymocytes are sensitive to relative lev- by guest on October 1, 2021 els of p56lckF505, consistent with the idea that signal “strength” Isolation of thymocyte subpopulations assists in determining the CD4ϩ vs CD8ϩ lineage choice. To isolate DP and CD4ϩ SP thymocytes for cell cycle analysis, total thymocytes were stained with fluorochrome-conjugated reagents specific for Materials and Methods CD4 and CD8 and sorted on FACStar (Becton Dickinson, San Jose, CA). ϩ ϩ Assembly of dLGF transgene construct and generation of For proliferation assay, the CD4 and CD8 T cells were purified from transgenic lines splenocytes by negative selection using magnetic bead separation methods. Briefly, splenocytes were incubated with Ab cocktail containing biotinyl- The lck gene containing the F505 mutation and a 0.6-kb fragment of the 3Ј ated anti-B220, anti-I-Ab, anti-CD11b, anti-NK1.1, and anti-CD4, or anti- human growth hormone gene was isolated from a previously reported con- CD8 monoclonal reagents (PharMingen, San Diego, CA) for 30 min and struct, pLGF (37), by restriction digest with NotI and StuI. The vector washed. Then magnetic beads preconjugated with streptavidin (Dynal, containing the lck distal promoter, pW120 (40), was linearized with BamHI Lake Success, NY) were added at an ϳ10:1 bead-target ratio and mixed for digest, filled in, and was subject to partial digestion with NotI to remove the 45 min at 4°C. Cells that did not bind the beads were collected after the human growth hormone gene. Finally, we inserted the lckF505 (including magnetic separation for experiments described in the text. the 3Ј-human growth hormone gene) gene into the lck distal promoter construct by directional ligation. The A-to-T point mutation converting the Cell cycle analysis tyrosine to a phenylalanine, and the joining region of the distal promoter and the lck gene was verified by nucleotide sequencing. Purified DP and SP thymocytes were stained with propidium iodide as The transgene DNA was microinjected into fertilized (C57BL/6 ϫ described by Hardy et al. (41) and analyzed by flow cytometry to determine nuclear DNA content. Briefly, sorted cells were washed in PBS, fixed in DBA/2)F2 embryos to generate transgenic founders. We determined transgene integration by Southern blot analysis of tail DNA and backcrossed 95% ethanol, and incubated in staining buffer containing propidium iodide four founders with C57BL/6 animals to establish transgenic lines. (Calbiochem, San Diego, CA) at 20 mg/ml, RNase A at 1 mg/ml, and 0.01% Nonidet P-40, followed by analysis on FACScan. Generation and screening of MHC-deficient dLGF animals Proliferation assay To generate MHC class I-deficient and MHC class II-deficient dLGF animals, we crossed dLGF animals from A16912 line with commercially Total or purified splenocytes were stimulated in RPMI (supplemented with ␤ ␤ Ϫ/Ϫ ␤bϪ/Ϫ available 2-microglobulin ( 2m) or I-A mice (Taconic Farms, 10% FCS, L-glutamine, nonessential amino acid, penicillin-streptomycin, Germantown, NY). To generate MHC double-deficient dLGF animals, we and 2-ME) containing various combinations of 145-2C11 ascites, PMA ␤ Ϫ/Ϫ ␤bϪ/Ϫ crossed the dLGF A16809 line animals with 2m /I-A mice, also (Calbiochem), recombinant murine IL-2 (Boehringer Mannheim, Indianap- purchased from Taconic Farms. olis, IN), and ionomycin (Calbiochem) for 2 days, pulsed with [3H]thymi- ␤ ␤b The genotype at the 2m and I-A loci was determined by PCR. The dine, and harvested on day 3. Proliferation was measured by the cpm in- ␤ primer sequences are as follows. For detection of wild-type 2m locus, the corporated and normalized to the number of mature T cells in each well. forward primer sequence is 5Ј-AAA CTG CAG TCA TCT TCC CCT GTG For allo-specific responses, indicated numbers of total splenocytes were GCC CTC AGA-3Ј, and the reverse primer sequence is 5Ј-GTA AGG cultured with irradiated splenocytes from C3H/HeJ animals in the presence AAG AAC TTG AGG CTT ACC-3Ј. For detection of wild-type I-A␤b or absence of exogenous IL-2. The culture was pulsed on day 4 and har- locus, the forward primer sequence is 5Ј-AGC ACC GCG CGG TGA CCG vested on day 5. The Journal of Immunology 2211

Results by a concomitant decrease in the percentage of CD4ϩCD8ϩ DP Generation of transgenic mouse lines and analyses of transgene thymocytes (from 87.0% to 34.7%). The increased proportion of expression SP thymocytes in dLGF animals reﬂected an increased production of mature T cells as indicated by the increased absolute number of To study the role of p56lck during late stages of thymocyte ontog- the SP thymocytes. Moreover, the number of SP thymocytes in- eny, we generated mice that express an activated form of p56lck creased as the abundance of p56lckF505 increased (Fig. 2B). Im- (p56lckF505) under the control of the lck distal promoter (Fig. 1A). These transgenic mice are herein referred to as dLGF mice. We portantly, the total number of thymocytes remained normal in all generated four independent dLGF lines by backcrossing founders dLGF lines, indicating that the expansion was selective for the SP compartment. At a 0.7:1 ratio (p56lckF505:wild-type p56lck in onto a C57BL/6 background. As shown in Fig. 1B, immunoblot lck analyses indicate that the total abundance of p56lck is increased in A16924 line), p56 F505 increased the number of SP thymocytes thymocytes from transgene-positive mice relative to littermate at least 3-fold relative to that of the control mice. Thus, production lck controls. Densitometric scanning of a representative blot showed of SP thymocytes is extremely sensitive to the level of p56 that transgene expression increased the total abundance of p56lck activity. ϳ1.4- to 2.5-fold (lanes 2, 4, 6, and 8) compared with littermate In contrast to thymocytes, the representation and the number of controls (lanes 1, 3, 5, and 7). The overall levels of transgene mature T cells in the periphery of dLGF mice are lower compared expression declined in the periphery (Fig. 1B, lanes 9–16), a result with the littermate control (Fig. 2A, lower panels). These data sug- lck different from the known pattern of lck distal promoter activity gest either that p56 F505 inhibits migration of mature thymo- lck (40). This discrepancy may be explained by the fact that the pro- cytes to the periphery or that the presence of p56 F505 is detri- portion of T cells among total splenocytes decreases in the periph- mental to the survival of peripheral T cells. Downloaded from ery of dLGF mice, as shown in Fig. 2A. In addition to the altered representation of thymocyte subsets, we observed that a large number of both the DP and the SP thy- lck Activated p56 increases representation of SP thymocytes mocytes in dLGF animals displayed reduced CD4 and CD8 core- dLGF animals appeared healthy and did not develop tumors of the ceptor expression on their cell surfaces (Fig. 2A, upper panels). thymi or peripheral lymphoid organs (up to 5 mo of age). We The mechanisms responsible for regulation of coreceptor expres-

assessed the effects of transgene expression on thymocyte devel- sion on T cell surfaces are presently unknown. Thymocytes from http://www.jimmunol.org/ opment by ﬂow cytometric analyses of CD4, CD8, and CD3 ex- dLGF animals also display reduced levels of surface CD3 proteins. pression. As shown in Fig. 2A, thymi from dLGF animals con- In control mice, ϳ10% of thymocytes express high levels of sur- tained substantially increased proportions of cells with mature face CD3 and correspond to SP mature cells (Fig. 2C, left). In CD4ϩCD8Ϫ and CD4ϪCD8ϩ phenotypes (upper panels). The in- contrast, only 2% of thymocytes from dLGF animals express high crease in the percentage of such SP thymocytes was accompanied levels of surface CD3 (right), even though the percentage of SP by guest on October 1, 2021

FIGURE 1. dLGF transgene and analysis of transgene expression. A, The dLGF transgene consists of the lck distal promoter, the lck gene containing a tyrosine-to-phenylalanine substitution at codon position 505, and a 0.6-kb fragment of the human growth hormone gene (hGH, includes poly(A) signal). B, Transgene expression in four independent lines was assessed by immunoblotting total lysates from thymocytes (lanes 1–8) and splenocytes (lanes 9–16) with anti-p56lck antisera. Control (C) and transgenic (Tg) littermates were analyzed for each line. Signal intensity was determined by densitometric analysis and the relative total intensity (transgenic/C) is indicated for each transgenic sample. 2212 p56lck-DRIVEN MATURATION OF SP THYMOCYTES

thymocytes is increased in these animals (10% in control vs 58% in transgenic mice). Moreover, the percentage of cells arbitrarily deﬁned as CD3mid is reduced (48.5% in control vs 23.7% in Tg animals) with a concomitant increase in the percentage of CD3Ϫ cells in dLGF animals, suggesting that a large proportion of DP thymocytes fail to up-regulate surface TCR-CD3 expression. Ac- tivation of p56lck has been previously correlated with decreased TCR-CD3 surface expression in immature and mature T cells. In DN thymocytes, activated p56lck, expressed under the control of the lck proximal promoter, inhibits TCR ␤-chain gene rearrangement and thus prevents production and surface expression of func- tionally rearranged TCR ␤-chains (42). In mature T cells, p56lckF505 down-regulates surface accumulation of TCR by di- recting TCR-CD3 complexes to lysosomal compartments for degradation (43). To determine which of these mechanisms may be responsible for the reduced accumulation of surface TCR/CD3 in dLGF animals, we permeabilized thymocytes from control, pLGF, and dLGF animals and stained them for cytoplasmic TCR ␤-chain (Fig. 2D). The percentage of cells containing cytoplasmic TCR ␤ protein is reduced in the DN thymocytes (Fig. 2D, center) from Downloaded from mice that express high levels of p56lckF505 under the control of the lck proximal promoter (pLGF (37)). These results are consistent with the known effects of p56lckF505 on TCR ␤-chain gene rearrangement. In contrast, the percentage of DN cells containing cytoplasmic ␤-chains remains normal in dLGF animals (Fig. 2D, right), indicating that TCR gene rearrangement is not inhibited. http://www.jimmunol.org/ Production of TCRs, judged by the content of intracellular ␤-chain proteins, is also normal in DP thymocytes from these mice (data not shown). This ﬁnding, together with the fact that p56lckF505 is not expressed in DN thymocytes (data not shown), is consistent with the idea that TCR-CD3 levels decrease in dLGF animals due to posttranslational degradation rather than diminished synthesis of functional TCR chains.

p56lckF505 does not promote cellular proliferation in dLGF by guest on October 1, 2021 animals The ﬁnding that p56lckF505 increases the proportion and the number of SP thymocytes suggests either that the transition from the DP to the SP stage is enhanced or that thymocytes in the SP compartment undergo proliferation. To address these possibilities, we puriﬁed DP and CD4ϩ SP thymocytes from control and dLGF animals and measured their DNA content. As shown in Fig. 3, the ϩ percentages of cells that are in S G2-M phase are comparable between Tg and control animals in both the DP and the SP com-

partments, and the percentage distribution of cells in G1 and in FIGURE 2. Expression of the dLGF transgene enhances production of ϩ S G2-M phases in our experiments is in agreement with previous SP thymocytes. A, Thymocytes (upper panels) and splenocytes (lower pan- reports (44). These results indicate that p56lckF505, at expression els) from a littermate control (left) and a transgenic animal (right) from the levels achieved in our system, does not induce spontaneous thy- highest expressing line (A16809) were stained with fluorochrome-conju- mocyte proliferation, arguing against the hypothesis that the in- gated mAbs specific for CD4 or CD8 molecules and analyzed by flow cytometry. Total thymocyte numbers and percentage of the live-gated pop- crease in SP thymocyte number results from postselection ulations are shown. Profiles are representative of three independent exper- proliferation. iments. B, Thymocyte numbers, expressed as fold over control values, are p56lckF505 drives differentiation of SP thymocytes plotted against relative total abundance of p56lck as determined by immunoblotting. The mean values for total thymocytes (छ) and SP thymocytes The observation that p56lckF505 increases the proportion and the ϩ ϩ (F, CD4 and CD8 combined) are shown with the SEM. Solid line, number of SP thymocytes, without inducing their proliferation, control value (NLC, normal littermate control), arbitrarily designated as 1. C, suggests that activated p56lck accelerates the maturation of thymo- Transgene expression diminishes surface CD3 levels on thymocytes. CD3 pro- cytes from the DP to the SP stage. In normal animals, maturation files of live cell-gated thymocytes are shown as single-parameter histograms. through this differentiation step requires the interaction of func- Percentages of CD3mid and CD3high populations are also indicated. D, Expres- tional TCRs with appropriate MHC molecules. However, we ob- sion of dLGF transgene does not inhibit early steps of TCR production. Thy- mocytes from control, pLGF, and dLGF animals were stained for surface CD4 served that SP thymocytes in dLGF animals develop efficiently and CD8, followed by permeabilization and staining for cytoplasmic TCR, and despite the reduced surface levels of CD3, suggesting that lck analyzed by three-color flow cytometry. Cytoplasmic TCR profiles of gated p56 F505 might compensate for any loss or attenuation of signals CD4ϪCD8Ϫ DN thymocytes are shown. Numbers indicate the percentages of normally triggered by TCR-MHC interaction. To test this possi- cells that stain positive for cytoplasmic TCR. bility, we investigated whether SP thymocytes can develop in The Journal of Immunology 2213

FIGURE 3. Cell cycle analysis. DNA content of thymocytes from dLGF (f) and control (o) animals were determined by propidium iodide staining and ﬂow cytometric analysis. The percentages of cells in G1 or S ϩ ϩ G2-M phase among sorted DP and CD4 SP populations are shown. The results are representative of three independent experiments.

␤ Ϫ/Ϫ Downloaded from dLGF mice bred to MHC class I ( 2m )-, MHC class II (I- A␤bϪ/Ϫ)-, or MHC double-deficient backgrounds. Fig. 4A shows CD4 and CD8 staining profiles of thymocytes from dLGF mice crossed onto an MHC class IϪ/Ϫ background. As expected, nontransgenic (non-Tg) MHC class IϪ/Ϫ mice (lower left) exhibited a profound deficiency of CD8ϩ SP thymocytes when compared with MHC class Iϩ/Ϫ mice (upper left). In contrast, expression of http://www.jimmunol.org/ p56lckF505 in MHC class IϪ/Ϫ mice (lower right) reconstitutes the CD8 SP compartments to levels comparable with that of its MHC class Iϩ/Ϫ/dLGF counterpart. These results indicate that p56lckF505 can drive the maturation of CD8ϩ SP thymocytes in the absence of MHC class I molecules. As in a wild-type background (Fig. 2C), the presence of the dLGF transgene caused decreased levels of surface TCR-CD3 on DP as well as on SP thymocytes in MHC class IϪ/Ϫ/dLGF mice (Fig. 4B, bottom). Thus, in by guest on October 1, 2021 dLGF mice (center and bottom), Ͻ10% of total CD8ϩ SP thymocytes express appreciable levels of TCR-CD3, and the level of TCR-CD3 expression in these cells is significantly reduced compared with MHC class Iϩ/Ϫ/non-Tg controls (top). Nonetheless, CD8ϩ SP thymocytes in MHC class IϪ/Ϫ/dLGF mice up-regulate CD69 as well as do their MHC class Iϩ/Ϫ/nontransgenic and MHC class Iϩ/Ϫ/dLGF counterparts (Fig. 4C). This suggests that dLGF thymocytes have received an activating signal during the DP-to-SP transition, even in the absence of MHC class I molecules. FIGURE 4. Expression of the dLGF transgene drives differentiation of lck Similarly, expression of p56 F505 in an MHC class II-defi- CD8ϩ SP thymocytes in a MHC class I-deficient environment. A, CD4 and ϩ cient background drove maturation of CD4 T cells. As shown in CD8 expression patterns of thymocytes from progeny of the dLGF ϫ MHC Ϫ Ϫ Fig. 5A, MHC class II / /dLGF mice possess a large number of class IϪ/Ϫ crosses were analyzed by flow cytometry. Profiles of thymocytes CD4ϩ SP thymocytes (lower right), in contrast to the MHC class from MHC class Iϩ/Ϫ/non-Tg (upper left), MHC class Iϩ/Ϫ/dLGF (upper Ϫ Ϫ Ϫ Ϫ IIϪ/Ϫ/nontransgenic control (lower left). We also noted that the right), MHC class I / /non-Tg (lower left), and MHC class I / /dLGF percent of CD4ϩ SP thymocytes in MHC class IIϪ/Ϫ/dLGF mice animals (lower right) are shown. Numbers indicate percentages of cells in was slightly reduced compared with the MHC class IIϩ/Ϫ/dLGF corresponding quadrants adjusted to capture discrete populations for each profile. Similar results were obtained from three independent experiments. control. This result suggests that MHC class II molecules can en- B, CD3 profiles of thymocytes are shown as single-parameter histograms. hance but are not required for maturation of SP thymocytes di- ϩ lck Solid lines, CD3 profiles of CD8 SP thymocytes; broken lines, DP thy- rected by p56 F505. The level of surface TCR/CD3 is again re- mocytes from animals with the indicated genotypes. Percentages of CD3ϩ Ϫ/Ϫ duced on thymocytes from MHC class II /dLGF mice (data not thymocytes are also shown. C, CD69 profiles for CD8ϩ SP (solid line) and ϩ shown), although CD4 SP thymocytes clearly up-regulate surface DP (broken line) thymocytes. The percentages of CD69ϩ population are expression of CD69 (data not shown) and the mature T cell marker indicated for the CD8ϩ SP thymocytes from each animal. Qa-2 (Fig. 5B). To test whether CD4ϩ T cells acquire the ability to respond to activating signals, we measured the ability of CD4ϩ T cells to proliferate. We purified CD4ϩ splenic T cells from MHC stimulators in the absence of exogenous IL-2, but those from MHC class II-positive (MHC class IIϩ/Ϫ) or -negative (MHC class class IIϪ/Ϫ/non-Tg, MHC class IIϩ/Ϫ/dLGF, and MHC class IIϪ/Ϫ) and nontransgenic or dLGF animals and cultured them with IIϪ/Ϫ/dLGF animals do not proliferate (Fig. 5C, left). Addition of H-2k allogeneic stimulators in the absence or presence of exoge- exogenous IL-2 enhances the overall response but still fails to nous IL-2. The CD4ϩ cells from MHC class IIϩ/Ϫ/nontransgenic restore proliferation of T cells from MHC class IIϩ/Ϫ/dLGF and control animals proliferate in response to increasing numbers of MHC class IIϪ/Ϫ/dLGF animals to the control level (Fig. 5C, 2214 p56lck-DRIVEN MATURATION OF SP THYMOCYTES

right). The apparent deficit in proliferative responses by MHC class IIϩ/Ϫ/dLGF and MHC class IIϪ/Ϫ/dLGF splenocytes may reflect the reduced levels of surface TCR on these cells (data not shown). This hypothesis is supported by the fact that stimulation with ionomycin and PMA, which bypasses TCR ligation, restores proliferation to near normal levels (Fig. 5D). The data above strongly suggest that p56lckF505 compensates for both the lack of appropriate MHC and the reduced TCR/CD3 levels on developing thymocytes, to drive lineage commitment and differentiation of thymocyte in vivo. To determine whether p56lckF505 is sufficient for both of these processes, we crossed dLGF mice onto an MHC-deficient (class I- and class II-deficient) background. Remarkably, large numbers of both CD4 and CD8 SP thymocytes develop in MHC-deficient dLGF mice (Fig. 6A, lower right). Importantly, the overall representation of thymocyte subsets remains unchanged compared with the MHC-positive dLGF control (upper right). Moreover, CD4 and CD8 SP thymocytes from MHC-deficient dLGF mice down-regulate HSA as efficiently as do their MHC-positive counterparts and better than the MHC-positive Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 5. Expression of the dLGF transgene drives differentiation of CD4ϩ SP thymocytes in MHC class II-deficient environment. A, CD4 and CD8 profiles of thymocytes from progeny of the dLGF ϫ MHC class IIϪ/Ϫ cross are shown. Numbers indicate percentages of thymocytes in corresponding quadrants. The results are representative of four independent experiments. B, The CD4ϩ SP thymocytes from MHC class IIϪ/Ϫ/dLGF animals express Qa-2. Surface expression of Qa-2 was analyzed by flow cytometry. In each panel, the solid line represents the Qa-2 profile of the FIGURE 6. Expression of dLGF transgene drives differentiation of SP ϩ CD4 SP thymocytes, and the broken line represents the profile of the DP thymocytes independent of MHC. A, CD4 and CD8 profiles of thymocytes ϩ thymocytes. Percentages of Qa-2 cells are indicated. C, Proliferation of from MHC-positive (MHCϩ/Ϫ ϭ class Iϩ/Ϫ/class IIϩ/Ϫ) non-Tg (upper ϩ ϩ purified CD4 T cells. Purified CD4 splenic T cells were cultured with left) and dLGF (upper right) animals and MHC-deficient (MHCϪ/Ϫ ϭ irradiated splenocytes from C3H (H-2k) animals. The culture was pulsed class IϪ/Ϫ/class IIϪ/Ϫ) counterparts (lower panels) are shown. The percent- with [3H]thymidine on day 4 and harvested on day 5. cpm are shown as the age of cells in each quadrant is also indicated. These profiles represent two mean of triplicate samples; error bars, SEM. The results from two MHC independent experiments. B, Expression of HSA was determined by triple- ϩ Ϫ ϩ Ϫ class II / /non-Tg animals (Ⅺ), one MHC class II / /dLGF animal (f), staining thymocytes with fluorochrome-conjugated reagents specific for Ϫ Ϫ Ϫ Ϫ one MHC class II / /non-Tg animal (E), and three MHC class II / / CD4, CD8, and HSA, followed by flow cytometric analysis. Solid line, ϩ dLGF animals (Q) are shown. D, Purified CD4 T cells were stimulated in HSA profile of indicated (CD4ϩ or CD8ϩ) SP thymocyte population; bro- culture with media only or with ionomycin and PMA. The culture was ken line, profile of the gated DP thymocytes for each animal. The numbers pulsed with [3H]thymidine on day 2 and harvested and counted on day 3. correspond to the percentage distribution of cells that have down-regulated The mean of triplicate wells and the SEM are also shown. or maintained high level expression of HSA. The Journal of Immunology 2215 non-Tg control (Fig. 6B). These results further support the hypoth- cells in the SP compartment to a ratio of ϳ10:1. Thereafter, further esis that p56lckF505 substitutes for a TCR-derived signal for pos- increases in p56lck abundance results in gradual normalization of itive selection and drives thymocyte differentiation to the mature the CD4:CD8 ratio. SP stage.

Relative p56lck activity determines the ratio of CD4:CD8 SP Discussion thymocytes Our study demonstrates that thymocytes are exquisitely sensitive We observed that the ratio of CD4:CD8 SP thymocytes varied to changes in p56lck activity during the transition from the DP to among the four dLGF lines. As depicted in Fig. 7A, the mean SP stage. The lck distal promoter-driven expression of p56lckF505 CD4:CD8 ratio of individual transgenic lines increases with in- at ϳ2.5- to 3-fold over endogenous wild-type p56lck levels causes creasing total p56lck abundance (A16924 and A16912 lines) but a 5-fold increase in the number of SP thymocytes (Fig. 1D). This abruptly returns to the normal ratio at the highest level of transgene result is in marked contrast to the effects of p56lck F505 expression expression (A16809 line). The ratios in individual animals from under the control of the lck proximal promoter (pLGF), which the two intermediate lines (A16912 and A16924) varied signifi- blocks generation of mature SP thymocytes (38). Several lines of cantly but were consistently higher than those in their littermate evidence strongly suggest that activated p56lck drives differentia- controls. We hypothesized that quantitative changes in total p56lck tion of mature SP thymocytes in dLGF mice. First, cell cycle anal- activity may directly influence the CD4:CD8 ratio of developing yses indicate that the proportion of proliferating cells in the DP and thymocytes or that the site of transgene integration indirectly af- SP compartments did not increase in dLGF mice (Fig. 3). Thus, fects the balance of lineage commitment. To test whether alter- massive expansion of SP thymocytes cannot easily explain the Downloaded from ations in p56lck abundance affect the CD4:CD8 ratio, we inter- increased number and representation of these cells. Second, SP crossed transgenic mice from the A16912 line expressing an thymocytes from dLGF animals express mature T cell markers: intermediate level of p56lckF505 (and exhibiting increased CD4: CD69 and Qa-2 are both induced, HSA is down-regulated, and CD8 ratios), thus generating animals hemizygous or homozygous mutually exclusive expression of the CD4 and CD8 coreceptors for the dLGF transgene. We analyzed three litters of animals from takes place. These results suggest that the presence of p56lckF505 this cross to individually assess the total p56lck abundance in thy- permits an increased number of DP thymocytes to survive and mocytes and to document the CD4:CD8 SP thymocyte ratio. The undergo final maturation in the thymus. http://www.jimmunol.org/ results, summarized in Fig. 7B, demonstrate that augmented Interestingly, surface TCR-CD3 and coreceptor expression is p56lckF505 levels initially increase the representation of CD4ϩ notably diminished in the DP and particularly in the SP compartment in dLGF animals. Our data show that production of TCR ␤-chains is normal in these mice. We further considered the pos- sibility that p56lckF505, acting as an agent that delivers all aspects of positively selecting signal, may cause premature cessation of the ␣-chain gene rearrangement in dLGF mice by shutting off the recombinase-activating gene 1 (RAG1) and RAG2 genes (45–48). by guest on October 1, 2021 However, TCR ␣-chain production occurs normally in dLGF mice, as judged by the level of endogenous ␣-chains expressed in dLGF- TCR double-transgenic mice (J. Alberola-Ila, personal communi- cation). Previous studies indicate that activated p56lck can down- regulate surface TCR by a posttranslational mechanism. Ab- mediated ligation of coreceptors, which presumably activates p56lck, decreases surface TCR expression in DP thymocytes (49, 50), whereas transgenic expression of dominant-negative p56lck increases surface TCR expression (39). Subsequently, expression of activated p56lck (p56lckF505) was correlated with trafficking of TCR-CD3 complexes to lysosomal compartment for degradation (43). Given that TCR-␤ protein levels are normal in DN thymocytes and that TCR ␣-chain production is normal, posttranslational degradation is almost certainly responsible for the reduced surface expression of TCRs in dLGF animals. It is also possible that, in the presence of activated p56lck, cells expressing fewer TCRs on their surfaces are preferentially selected and survive. However, down- regulation of the TCR clearly takes place in the DP compartment (before selection), indicating that the catalytically active p56lck can directly down-modulate surface expression of this receptor. The effects of p56lck activation on coreceptor expression remain unclear. However, in vivo administration of anti-CD4 mAbs has been shown to diminish CD4 expression in addition to its effect on lck FIGURE 7. CD4/CD8 ratio and total p56 activity. A, The ratio of surface TCR expression (49), suggesting that p56lck may play a CD4ϩ SP thymocytes to the CD8ϩ SP thymocytes (CD4:CD8) was cal- role in regulation of coreceptor expression or that expression of culated for the four dLGF-transgenic lines and their non-Tg littermates. The mean ratio values are plotted against relative total p56lck abundance, as coreceptors is in some way coordinately regulated with that of determined by immunoblotting and densitometry. B, The CD4:CD8 ratios TCR-CD3. for individual progeny from intercrossing the A16912 line animals are Current models of thymocyte selection posit that the avidity of plotted against relative total p56lck abundance. The results summarize the the combined interactions among the TCR, coreceptor, and the data from three independent litters. MHC influences the cell fate choice in one of three ways. If the 2216 p56lck-DRIVEN MATURATION OF SP THYMOCYTES combined interaction is too weak (or does not occur), cells die by T cell activation has been previously correlated with anergy neglect. If the interaction is too strong, cells die by negative se- (55). IL-2 production by purified CD4ϩ T cells from MHC class lection. In contrast, those cells that interact with intermediate af- IIϪ/Ϫ/dLGF mice was indeed greatly diminished in our exper- finity receive the appropriate signal for survival and differentiation iments (data not shown), suggesting that anergy in vivo may at and become mature thymocytes. Manipulations that alter these sur- least partially account for the observed unresponsiveness of face interactions directly affect the efficiency with which mature T these cells in vitro. Finally, the TCR repertoire in dLGF mice lck cells are generated. The fact that p56 F505 increases the number may be less sensitive to distinguishing between self and nonself of SP thymocytes suggests that cells interpret the presence of because the constitutively activated p56lck drives thymocyte lck p56 F505 as having received an optimal signal for survival/dif- differentiation regardless of individual TCR specificity. In this lck ferentiation. Quantitatively, this may mean that p56 F505 in- scenario, relative frequencies of self-restricted T cells would be creases signal strength to permit survival of cells that would nor- reduced, accompanied by an increased production of potentially lck mally die by neglect. Whether p56 F505 also increases the self-reactive T cells. In support of this hypothesis, CD8ϩ cy- number of cells that die in negative selection was not addressed in totoxic T cells from dLGF mice exhibited elevated levels of this study. Nonetheless, the augmentation of signal strength cross-reactivity toward syngeneic targets in vitro, relative to achieved by modest expression of the dLGF transgene (encoding a non-Tg controls (data not shown). 7-fold more active kinase protein (51) clearly biases the tendency We also present data that suggest that total p56lck activity can toward survival than death, because the number of SP thymocytes influence the relative representation of CD4ϩ vs CD8ϩ SP thy- increases as a direct function of the relative level of transgene lck lck mocyte population. Expression of p56 F505 increased the repre- expression (Fig. 2B). Alternatively, threshold activation of p56 ϩ may be the qualitative requirement for survival and/or differenti- sentation of CD4 SP thymocytes (Fig. 7A), a trend that was re- Downloaded from ation of thymocytes undergoing selection. If this were true, all verted by intercrossing the dLGF mice to further increase in the cells that possess p56lckF505 may potentially survive and mature. abundance of transgene protein (Fig. 7B). We speculate that at low lck The tyrosine-to-phenylalanine mutation renders p56lck constitu- levels, p56 F505 promotes differentiation and/or survival of CD4 tively active, which has been shown to greatly potentiate Ag re- lineage cells as a result of its preferred interaction with CD4 (28). lck ceptor-mediated signals in T cells (32). Thus, we tested whether At higher levels of p56 F505, the CD4:CD8 ratio returns to nor- lck the p56lckF505-driven differentiation process required TCR liga- mal as the p56 F505-CD4 interactions become saturated and http://www.jimmunol.org/ lck lck tion at all, by breeding dLGF animals directly onto MHC class I-, p56 F505-CD8 interactions increase. Whether p56 F505 sends class II-, or double-deficient background. Our data show that an instructive signal to direct lineage commitment or rather pro- p56lckF505 drives both maturation of CD8ϩ SP thymocytes in the motes survival of precommitted thymocytes remains unclear. Pre- absence of MHC class I and maturation of CD4ϩ SP thymocytes viously, a model was proposed in which the relative strength of the in the absence of MHC class II, implying that constitutively acti- signal generated by coreceptor-associated p56lck determines CD4 vated p56lckF505 can direct thymocyte maturation in the absence and CD8 lineage preference (24). The quantitative interpretation of of appropriate MHC engagement. It is unlikely that the effect of this model and our current data is consistent with the observation p56lckF505 expression is to promote aberrant selection of SP cells that p56lck interacts with the cytoplasmic tail of the CD4 molecule ϩ ϩ by guest on October 1, 2021 on mismatched MHC molecule (52) because CD4 and CD8 SP with higher affinity than it does with the CD8 molecule (28). Other thymocytes emerge even in MHC double-deficient dLGF animals. studies have indicated that thymocyte lineage commitment occurs These results strengthen our view that p56lckF505 drives survival independent of TCR-MHC specificity and even independent of and maturation of thymocytes independent of TCR ligation by coreceptors, before selection (17, 19, 20). In our system, the de- lck MHC. The ability of activated p56 to substitute for pre-TCR- ficiency of MHC exerted little effect on the ability of p56lckF505 to dependent signals has been previously demonstrated. In RAG-1- influence CD4:CD8 ratios (Figs. 4A,5A, and 6A), suggesting that deficient mice crossed with pLGF-transgenic animals, expression the total activity of p56lck, rather than coreceptor ligation (by lck of p56 F505 (under the control of lck proximal promoter) was MHC), determines lineage choice and/or survival. Two recently sufficient to overcome the block at the DN-to-DP transition (53). published studies further strengthen our view that total p56lck ac- Our results demonstrate that the presence of an activated form of tivity determines the CD4:CD8 ratios. In one study, the level of p56lck can bypass the requirement for TCR ligation during the total p56lck activity was shown to override MHC restriction of a DP-to-SP transition. The phenotype of dLGF mice is also similar given TCR in driving the lineage commitment decisions (56). In a to the phenotype of mice lacking Csk, a negative regulator of Ϫ Ϫ second study, reconstitution of wild-type p56lck in lck / mice p56lck (54). In these mice, the absence of Csk in immature thy- resulted in maturation predominantly of the CD4ϩ SP population mocytes drove thymocyte differentiation through the DP and SP (57). The observed skewing toward the CD4 lineage in this system stages, independent of TCR ligation. The data from our study im- is consistent with our data that increased p56lck activity initially ply that Csk deficiency leads to hyperactivation (or prolonged ac- ϩ lck favors differentiation of CD4 cells (Fig. 2B) and suggests that the tivation) of p56 , which in turn drives thymocyte maturation. lck Mature T cells that develop in dLGF animals exit the thymus level of activity achieved by the expression of the wild-type p56 and circulate in the periphery but exhibit limited functional capac- must fall within this range. A comprehensive series of experiments lck ity. Purified CD4ϩ T cells from MHC class IIϪ/Ϫ/dLGF animals now supports the view that p56 is a pivotal regulator of early proliferated only weakly in response to allogeneic stimulators (Fig. thymocyte development entrained by the pre-TCR. Our studies 5C), although stimulation with ionomycin and phorbol ester re- make plain that the selection processes ordinarily attributed to stored the response significantly (Fig. 5D). These results are per- TCR signaling later in thymocyte maturation can also be mimicked lck haps not surprising, given that T cells from dLGF mice possess by simple activation of p56 . This observation encompasses not substantially decreased levels of surface TCR, and these cells may only the emergence of SP cells with “mature” cell surface pheno- remain refractory to TCR engagement. Alternatively, the pres- types but also the relative proportion of thymocytes that display ϩ ϩ ence of p56lckF505 in mature T cells may be interpreted as TCR CD4 vs CD8 characteristics. It is plausible that the strength of engagement and, in the absence of appropriate costimulatory the p56lck-derived signal can by itself direct all aspects of lineage signal, may result in anergy. The lack of IL-2 production during commitment during thymocyte maturation. The Journal of Immunology 2217

Acknowledgments kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs. Cell 60:755. We thank Dr. Michael Bevan for helpful suggestions during the course of 29. Veillette, A., M. A. Bookman, E. M. Horak, L. E. Samelson, and J. B. Bolen. this study; and Drs. M. Bevan, Anne Norment, Steven Levin, Laurel Lenz, 1989. Signal transduction through the CD4 receptor involves the activation of the and Michael Farrar for critical review of the manuscript. internal membrane tyrosine-protein kinase p56lck. Nature 338:257. 30. Luo, K. X., and B. M. Sefton. 1990. Analysis of the sites in p56lck whose phos- phorylation is induced by tetradecanoyl phorbol acetate. Mol. Cell Biol. 10:5305. References 31. Marth, J. D., D. B. Lewis, M. P. Cooke, E. D. Mellins, M. E. Gearn, 1. von Boehmer, H. 1994. Positive selection of lymphocytes. Cell 76:219. L. E. Samelson, C. B. Wilson, A. D. Miller, and R. M. Perlmutter. 1989. Lym- 2. Jameson, S. C., K. A. Hogquist, and M. J. Bevan. 1995. Positive selection of phocyte activation provokes modification of a lymphocyte-specific protein ty- thymocytes. Annu. Rev. Immunol. 13:93. rosine kinase (p56lck). J. Immunol. 142:2430. 3. Teh, H. S., P. Kisielow, B. Scott, H. Kishi, Y. Uematsu, H. Bluthmann, and 32. Abraham, N., M. C. Miceli, J. R. Parnes, and A. Veillette. 1991. Enhancement of H. von Boehmer. 1988. Thymic major histocompatibility complex antigens and T-cell responsiveness by the lymphocyte-specific tyrosine protein kinase p56lck. the specificity of the ␣␤ T cell receptor determine the CD4/CD8 phenotype of T Nature 350:62. cells. Nature 335:229. 33. Glaichenhaus, N., N. Shastri, D. R. Littman, and J. M. Turner. 1991. Requirement 4. Sha, W. C., C. A. Nelson, R. D. Newberry, D. M. Kranz, J. H. Russell, and for association of p56lck with CD4 in antigen-specific signal transduction in T D. Y. Loh. 1988. Positive and negative selection of an antigen receptor on T cells cells. Cell 64:511. in transgenic mice. Nature 336:73. 34. Straus, D. B., and A. Weiss. 1992. Genetic evidence for the involvement of the 5. Berg, L. J., A. M. Pullen, B. Fazekas de St. Groth, D. Mathis, C. Benoist, and lck tyrosine kinase in signal transduction through the T cell antigen receptor. Cell M. M. Davis. 1989. Antigen/MHC-specific T cells are preferentially exported 70:585. from the thymus in the presence of their MHC ligand. Cell 58:1035. 35. Molina, T. J., K. Kishihara, D. P. Siderovski, W. van Ewijk, A. Narendran, 6. Kaye, J., M. L. Hsu, M. E. Sauron, J. C. Jameson, R. J. Gascoigne, and E. Timms, A. Wakeham, C. J. Paige, K. U. Hartman, and A. Veillette. 1992. S. M. Hedrick. 1989. Selective development of CD4ϩ T cell in transgenic mice Profound block in thymocyte development in mice lacking p56lck. Nature 357: expressing a class II MHC-restricted antigen receptor. Nature 341:746. 161. 7. Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, and R. Jaenisch. 1990. 36. Levin, S. D., S. J. Anderson, K. A. Forbush, and R. M. Perlmutter. 1993. A ␤ Ϫ ϩ lck 2-microglobulin deficient mice lack CD4 8 cytolytic T cells. Nature 344:742. dominant-negative transgene defines a role for p56 in thymopoiesis. EMBO J 8. Koller, B. H., P. Marrack, J. W. Kappler, and O. Smithies. 1990. Normal devel- 12:1671. Downloaded from ␤ ϩ opment of mice deficient in 2M, MHC class I proteins, and CD8 T cells. 37. Abraham, K. M., S. D. Levin, J. D. Marth, K. A. Forbush, and R. M. Perlmutter. Science 248:1227. 1991. Delayed thymocyte development induced by augmented expression of 9. Cosgrove, D., D. Gray, A. Dierich, J. Kaufman, M. Lemeur, C. Benoist, and p56lck. J. Exp. Med. 173:1421. D. Mathis. 1991. Mice lacking MHC class II molecule. Cell 66:1051. 38. Abraham, K. M., S. D. Levin, J. D. Marth, K. A. Forbush, and R. M. Perlmutter. 10. Grusby, M. J., R. S. Johnson, V. E. Papaioannou, and L. H. Glimcher. 1991. 1991. Thymic tumorigenesis induced by overexpression of p56lck. Proc. Natl. Depletion of CD4ϩ T cells in major histocompatibility complex class II-deficient Acad. Sci. USA 88:3977. mice. Science 253:1417. 39. Hashimoto, K., S. J. Sohn, S. D. Levin, T. Tada, R. M. Perlmutter, and lck 11. Fung-Leung, W. P., M. W. Schilham, A. Rahemtulla, T. M. Kundig, T. Nakayama. 1996. Requirement for p56 tyrosine kinase activation in T cell http://www.jimmunol.org/ M. Vollenweider, J. Potter, W. van Ewijk, and T. W. Mak. 1991. CD8 is needed receptor-mediated thymic selection. J. Exp. Med. 184:931. for development of cytotoxic T cells but not helper T cells. Cell 65:443. 40. Wildin, R. S., H. U. Wang, K. A. Forbush, and R. M. Perlmutter. 1995. Func- 12. Nakayama, K., I. Negishi, K. Kuida, M. C. Louie, O. Kanagawa, H. Nakauchi, tional dissection of the murine lck distal promoter. J. Immunol. 155:1286. and D. Y. Loh. 1994. Requirement for CD8 ␤ chain in positive selection of 41. Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, and K. Hayakawa. 1991. CD8-lineage T cells. Science 263:1131. Resolution and characterization of pro-B and pre-pro-B cell stages in normal 13. Crooks, M. E. C., and D. R. Littman. 1994. Disruption of T lymphocyte positive mouse bone marrow. J. Exp. Med. 173:1213. and negative selection in mice lacking the CD8 ␤ chain. Immunity 1:277. 42. Anderson, S. J., K. M. Abraham, T. Nakayama, A. Singer, and R. M. Perlmutter. 14. Rahemtulla, A., W. P. Fung-Leung, M. W. Schilham, T. M. Kundig, 1992. Inhibition of T-cell receptor ␤-chain gene rearrangement by overexpression S. R. Sambhara, A. Narendran, A. Arabian, A. Wakeham, C. J. Paige, of the non-receptor protein tyrosine kinase p56lck. EMBO J 11:4877. R. M. Zinkernagel, et al. 1991. Normal development and function of CD8ϩ cells 43. D’Oro, U., M. S. Vacchio, A. M. Weissman, and J. D. Ashwell. 1997. Activation but markedly decreased helper cell activity in mice lacking CD4. Nature 353:180. of the Lck tyrosine kinase targets cell surface T cell antigen receptors for lyso-

15. von Boehmer, H. 1996. CD4/CD8 lineage commitment: back to instruction? somal degradation. Immunity 7:619. by guest on October 1, 2021 J. Exp. Med. 183:713. 44. Egerton, M., R. Scollay, and K. Shortman. 1990. Kinetics of mature T-cell de- 16. Robey, E., A. Itano, W. C. Fanslow, and B. J. Fowlkes. 1994. Constitutive CD8 velopment in the thymus. Proc. Natl. Acad. Sci. USA 87:2579. expression allows inefficient maturation of CD4ϩ helper T cells in class II major 45. Borgulya, P., H. Kishi, Y. Uematsu, and H. von Boehmer. 1992. Exclusion and histocompatibility complex mutant mice. J. Exp. Med. 179:1997. inclusion of ␣ and ␤ T cell receptor alleles. Cell 69:529. 17. Corbella, P., D. Moskophidis, E. Spanopoulou, C. Mamalaki, M. Tolaini, 46. Brandle, D., C. Muller, T. Rulicke, H. Hengartner, and H. Pircher. 1992. En- A. Itano, D. Lans, D. Baltimore, E. Robey, and D. Kioussis. 1994. Functional gagement of the T cell receptor during positive selection in the thymus down- commitment to helper T cell lineage precedes positive selection and is indepen- regulates RAG-1 expression. Proc. Natl. Acad. Sci. USA 89:9529. dent of T cell receptor MHC specificity. Immunity 1:269. 47. Petrie, H. T., F. Livak, D. G. Schatz, A. Strasser, I. N. Crispe, and K. Shortman. 18. Davis, C. B., N. Killeen, M. E. C. Crooks, D. Raulet, and D. R. Littman. 1993. 1993. Multiple rearrangements in T cell receptor ␣ chain genes maximize the Evidence for a stochastic mechanism in the differentiation of mature subsets of T production of useful thymocytes. J. Exp. Med. 178:615. lymphocytes. Cell 73:237. 48. Kouskoff, V., J.-L. Vonesch, C. Benoist, and D. Mathis. 1995. The influence of 19. Goldrath, A. W., K. A. Hogquist, and M. J. Bevan. 1997. CD8 lineage commit- positive selection on RAG expression in thymocytes. Eur. J. Immunol. 25:54. ment in the absence of CD8. Immunity 6:633. 49. McCarthy, S. A., A. M. Kruisbeek, I. K. Uppenkamp, S. O. Sharrow, and 20. Sebzda, E., M. Choi, W. P. Fung-Leung, T. W. Mak, and P. S. Ohashi. 1997. A. Singer. 1988. Engagement of the CD4 molecule influences cell surface ex- Peptide-induced positive selection of TCR transgenic thymocytes in a coreceptor- pression of the T-cell receptor on thymocytes. Nature 336:76. independent manner. Immunity 6:643. 50. Bonifacino, J. S., S. A. McCarthy, J. E. Maguire, T. Nakayama, D. S. Singer, 21. Irving, B. A., and A. Weiss. 1991. The cytoplasmic domain of the T cell receptor R. D. Klausner, and A. Singer. 1990. Novel post-translational regulation of TCR ␨ chain is sufficient to couple to receptor-associated signal transduction pathways. expression in CD4ϩCD8ϩ thymocytes influenced by CD4. Nature 344:247. Cell 64:891. 51. Marth, J. D., J. A. Cooper, C. S. King, S. F. Ziegler, D. A. Tinker, R. W. Overell, 22. Wegener, A. K., F. Letourneur, A. Hoeveler, T. Brocker, F. Luton, and E. G. Krebs, and R. M. Perlmutter. 1988. Neoplastic transformation induced by B. Malissen. 1992. The T cell receptor/CD3 complex is composed of at least two an activated lymphocyte-specific protein tyrosine kinase (pp56lck). Mol. Cell Biol. autonomous transduction modules. Cell 68:83. 8:540. 23. Haughn, L., S. Gratton, L. Caron, R. Sekaly, A. Veillette, and M. Julius. 1992. 52. Kirberg, J., A. Baron, S. Jakob, A. Rolink, K. Karjalainen, and H. von Boehmer. Association of tyrosine kinase p56lck with CD4 inhibits the induction of growth 1994. Thymic selection of CD8ϩ single positive cells with a class II major his- through the ␣␤ T-cell receptor. Nature 358:328. tocompatibility complex-restricted receptor. J. Exp. Med. 180:25. 24. Itano, A., P. Salmon, D. Kioussis, M. Tolanis, P. Corbella, and E. Robey. 1996. 53. Mombaerts, P., S. J. Anderson, R. M. Perlmutter, T. W. Mak, and S. Tonegawa. The cytoplasmic domain of CD4 promotes the development of CD4 lineage T 1994. An activated lck transgene promotes thymocyte development in RAG-1 cells. J. Exp. Med. 183:731. mutant mice. Immunity 1:261. 25. Perlmutter, R. M., S. D. Levin, M. W. Appleby, S. J. Anderson, and 54. Schmedt, C., K. Saijo, T. Niidome, R. Kuhn, S. Aizawa, and A. Tarakhovsky. J. Alberola-Ila. 1993. Regulation of lymphocyte function by protein phosphory- 1998. Csk controls antigen receptor-mediated development and selection of T- lation. Annu. Rev. Immunol. 11:451. lineage cells. Nature 394:901. 26. Veillette, A., M. A. Bookman, E. M. Horak, and J. B. Bolen. 1988. The CD4 and 55. Lenardo, M. J. 1991. Interleukin-2 programs mouse ␣␤ T lymphocytes for apo- CD8 T cell surface antigens are associated with the internal membrane tyrosine- ptosis. Nature 353:858. protein kinase p56lck. Cell 55:301. 56. Hernandez-Hoyos, G., S. J. Sohn, E. V. Rothenberg, and J. Alberola-Ila. 2000. 27. Shaw, A. S., K. E. Amrein, C. Hammond, D. F. Stern, B. M. Sefton, and J. K. Rose. Lck activity controls CD4/CD8 T cell lineage commitment. Immunity 12:313. 1989. The lck tyrosine protein kinase interacts with the cytoplasmic tail of the CD4 57. Legname, G., B. Seddon, M. Lovatt, P. Tomlinson, N. Sarner, M. Tolaini, glycoprotein through its unique amino-terminal domain. Cell 59:627. K. Williams, T. Norton, D. Kioussis, and R. Zamoyska. 2000. Inducible expres- 28. Turner, J. M., M. H. Brodsky, B. A. Irving, S. D. Levin, R. M. Perlmutter, and sion of a p56Lck transgene reveals a central role for Lck in the differentiation of D. R. Littman. 1990. Interaction of the unique N-terminal region of tyrosine CD4 SP thymocytes. Immunity 12:537.