Analysis and Expression of a Cloned Pre-T Cell Receptor Gene Claude Saint-Ruf, Katarina Ungewiss, Markus Groettrup, Ludovica Bruno, Hans Fehling, Harald Von Boehmer

Analysis and expression of a cloned pre-T cell receptor gene Claude Saint-Ruf, Katarina Ungewiss, Markus Groettrup, Ludovica Bruno, Hans Fehling, Harald von Boehmer

To cite this version:

Claude Saint-Ruf, Katarina Ungewiss, Markus Groettrup, Ludovica Bruno, Hans Fehling, et al.. Analysis and expression of a cloned pre-T cell receptor gene. Science, American Association for the Advancement of Science, 1994, 266 (5188), pp.1208-1212. 10.1126/science.7973703. inserm- 02177273

HAL Id: inserm-02177273 https://www.hal.inserm.fr/inserm-02177273 Submitted on 8 Jul 2019

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. .M ... ..l ...

18. Russian Federation, statute 2667-1 (4 April 1992). 28. J. V. Jemski and G. B. Phillips, in Methods ofAnimal 38. H. A. Druett, Nature 170, 288 (1952). 19. L. Grinberg, personal communication. Experimentation, W. 1. Gay, Ed. (Academic Press, 39. R. Scherrer and V. E. Shull, Can. J. Microbiol. 33, 20. V. M. Popugaylo, R. P. Sukhanova, M. I. Kukhto, in New York, 1965), pp. 273-341. 304 (1987). Current Problems of Anthrax Prophylaxis in the 29. A. C. Chamberlain, Proc. R. Soc. London A 296, 45 40. We thank A. V. Yablokov, Counsellor to the Presi- USSR (Moscow, 1974), p. 50. (1967). dent of Russia for Ecology and Health, for letters of 21. S. R. Hanna, G. A. Briggs, R. P. Hosker, Handbook 30. T. W. Horst, in Atmospheric Sulfur Deposition, D. S. introduction; Ural State University and its then rec- on Atmospheric Diffusion (U.S. Department of Ener- Shriner, C. R. Richmond, S. E. Lindberg, Eds. (Ann tor, P. E. Suetin, for inviting us to Ekaterinburg; S. gy Report No. DOE/TIC-11223, Washington, DC, Arbor Science, Ann Arbor, Ml, 1980), pp. 275-283. F. Borisov, V. A. Shchepetkin, and A. P. Tiutiunnik 1982). 31. A. Bovallius and P. Anas, in Proceedings of the 1st for assistance and advice; members of the Ekater- 22. D. E. Davids and A. R. Lejeune, Secondary Aerosol International Conference on Aerobiology, Interna- inburg medical community and the Sverdlovskaya Hazard in the Field (Defence Research Establish- tional Association for Aerobiology, Munich, West Oblast Sanitary Epidemiological Service for discus- ment Suffield Report No. 321, Ralston, Alberta, Can- Germany, 13 to 15 August 1978, A. W. Frankland, E. sions, notes, and documents; interview respon- ada, 1981). Stix, H. Ziegler, Eds. (Erich Schmidt, Berlin, 1980), dents for their cooperation; P. N. Burgasov, USSR 23. A. Birenzvige, Inhalation Hazard from Reaerosolized pp. 227-231. Deputy Minister of Health at the time of the out- Biological Agents: A Review (U.S. Army Chemical 32. G. A. Cristy and C. V. Chester, Emergency Protec- break, for documents regarding livestock deaths; Research, Development and Engineering Center Re- tion Against Aerosols (Report No. ORNL-5519, Oak and People's Deputy L. P. Mishustina for the ad- port No. TR-413, Aberdeen, MD, 1992). Ridge National Laboratory, Oak Ridge, TN, July ministrative list of those who died. I. V. Belaeva 24. P. S. Brachman, S. A. Plotkin, F. H. Bumford, M. M. 1981). assisted with interviews. We also thank B. Ring, W. Atchison, Am. J. Hyg. 72, 6 (1960). 33. D. S. Ditmer and R. M. Grebe, Eds., Handbook of H. Bossert, P. J. M. Cannone, M. T. Collins, S. R. 25. D. W. Henderson, S. Peacock, F. C. Belton, J. Hyg. Respiration (Saunders, Philadelphia, 1958). Hanna, J. V. Jemski, D. Joseph, H. F. Judson, M. 54, 28 (1956); C. A. Gleiser, C. C. Berdjis, H. A. 34. G. A. Young, M. R. Zelle, R. E. Lincoln, J. Infect. Dis. M. Kaplan, J. Medema, C. R. Replogle, R. Stafford, Hartman, W. S. Gochenour, Br. J. Exp. Pathol. 44, 79, 233 (1946). J. H. Steele, and E. D. Sverdlov. Supported by 416 (1963). 35. J. V. Jemski, personal communication. grants to M.M. from the John D. and Catherine T. 26. H. N. Glassman, Bacteriol. Rev. 30, 657 (1966). 36. C. I. Bliss, Ann. Appl. Biol. 22,134 (1935). MacArthur Foundation and the Carnegie Corpora- 27. A. V. Elkina, Zh. Mikrobiol. Epidemiol. Immunobiol. 37. H. A. Druett, D. W. Henderson, L. Packman, S. J. tion of New York. This article is dedicated to Alex- 1971 (no. 9), 112 (1971). Peacock, J. Hyg. 51, 359 (1953). ander Langmuir.

80-kD complex suggested that either the Downloaded from * RESEARCH ARTICLE complex was a homodimer or that an un- known TCR chain was involved that may affect T cell maturation. A glycosylated chain of 33 kD (gp33) is paired with TCRP Analysis and Expression of a proteins in a TCR,-transfected immature T cell line (SCB.29) from severe combined http://science.sciencemag.org/ Cloned Pre-T Cell Receptor Gene immunodeficient (SCID) mice (12), but could not be identified in normal thymo- Claude Saint-Ruf,* Katharina Ungewiss,* Marcus Groettrup, cytes (12, 13). The gp33-TCR3 complex of Ludovica Bruno, Hans Joerg Fehling, Harald von Boehmer SCB.29 cells is associated with CD3 proteins (8, 12) and cross-linking of TCR chains initiates Ca2+ mobilization. This The T cell antigen receptor (TCR) P3 chain regulates early T cell development in the suggested that this TCR, complex could be absence of the TCRoL chain. The developmentally controlled gene described here responsible for the developmental progres- encodes the pre-TCRa (pTcx) chain, which covalently associates with TCRf and with sion observed in TCRP transgenic, rear-

the CD3 proteins forms a pre-TCR complex that transduces signals in immature rangement-deficient mice, whereas the on July 9, 2019 thymocytes. Unlike the X5 pre-B cell receptor protein, the pTcx chain is a type I TCRP GPI-linked monomer could repre- transmembrane protein whose cytoplasmic tail contains two potential phosphorylation sent a transgenic artifact (14, 15). We have sites and a Src homology 3 (SH3)-domain binding sequence. Pre-TCRa transfection now cloned the gene encoding gp33 and experiments indicated that surface expression of the pre-TCR is controlled by addi- examined its structure and expression. Be- tional developmentally regulated proteins. Identification of the pTa gene represents an cause of its properties, the gp33 protein was essential step in the structure-function analysis of the pre-TCR complex. named the pre-TCRa (pTax) chain. Pre-T cell receptor a (pTa) expression in immature T cells. The pTa chain can be identified by two-dimensional (diagonal) T cell development takes place in discrete chains, are sufficient to promote early T cell gel electrophoresis, in which the disulfide- steps during which the TCR genes are rear- development (6-8). Although such mice linked pTao protein under reducing condi- ranged and expressed in temporal order. are still rearrangement-defective, their im- tions migrates away from the diagonal just During development of TCRax-expressing mature thymocytes (which express neither underneath the TCR, protein (12) (Figs. 1 cells the TCRP gene is rearranged and ex- the CD4 nor CD8 proteins) begin to express and 2). The analytical method was scaled pressed before the TCRao gene (1, 2). With- CD4 and CD8 coreceptors, transcripts of the up to obtain sufficient amounts of pTa out TCR rearrangement the development of TCRcx locus become detectable (7), and the protein for microsequencing. In a first at- T cells is arrested at an early stage (3-5). By number of thymocytes increases (6-8). In- tempt a 20-amino-acid-long NH2-terminal introducing TCRP transgenes into mice troduction of TCRP transgenes into normal sequence was obtained; a peptide of the 18 that are defective for rearrangement of an- mice suppresses rearrangement of endoge- NH2-terminal residues was synthesized and tigen receptor genes, it was shown that nous TCRP genes (9, 10). The TCRI trans- injected into rabbits to obtain a pTa-spe- TCRP proteins, in the absence of TCRa gene is expressed on the cell surface in the cific antiserum. The antiserum was tested absence of TCRaL proteins in both normal for binding to the pTao protein. To this end C. Saint-Ruf is at the Unite INSERM 373, Institut (11) as well as in rearrangement-defective lysates from the TCRax-negative SCB.29 France. K. Necker, 75730 Paris, Cedex 15, Ungewiss, in an 80-kD disulfied-linked cell line as well as the TCRcx3-expressing M. Groettrup, L. Bruno, H. J. Fehling, and H. von Boe- mice (7, 8, 12) hmer are at the Basel Institute for Immunology, CH- complex and as a glycosyl-phosphatidylino- B6.2.16BW hybridoma (12) were precipi- 4005 Basel, Switzerland. sitol (GPI)-linked 40-kD monomer. tated with the monoclonal antibody (mAb) 'The first two authors contributed equally to this work. The presence of the TCRB chain in the F23.1 to V38 proteins (16). Precipitates

1208 SCIENCE * VOL. 266 * 18 NOVEMBER 1994 S ~ lail~

Fig. 1. Identification of the SCB.29 B6.2.16BW pTa chain by antibodies. Im- munoprecipitations of TCR3 RR from either SCB.29 or ~~~~~~~~~~~~~~~~~~~~~~~9 B6.2.16BW cells were im- 69- munoblotted with control re- 69 : 'r' : ~~~~~4545j : 69 t16= agents or antibodies to TCRI or pTot, with or with- ~~~30-

out peptide (PEP) from the 21 21 2121 NH2-terminus of pTa (38). SCB.29, cell line from SCID GantiR-HRPO F23.1.1 B F23.1.1 B+PEPF23.1.1 .B mouse transfected with TCRj3; B6.2.1 6BW, TCRao3- NR expressing T cell hybridoma; RV GantiR-HRPO, goat antibodies to rabbit immuno- 4-69t 69 -69- globulin conjugated to 45- 69 horseradish peroxidase; 45 45 45 F23.1, mAb to TCRB (V.8); 30 30 30- P153, pre-immune serum 21- from rabbit 53; S53, im- 21- 21 - 21 mune serum from rabbit 53 P153 S53 S53 + PEP S53 injected with NH2-terminal peptide of pTax; CP53, affinity column-purified S53. NR Downloaded from Molecular sizes are given in r kilodaltons. 69- 45 - 30fl 21f- http://science.sciencemag.org/

were separated on diagonal gels and the mocytes was not previously identified, we terminal protein sequence and an oligonu- proteins analyzed in Western immunoblots used the above approach to examine this cleotide complementary to the lambda ZAP by staining with either mAb F23.1 or serum complex in thymocytes from TCRa,-defi- vector-sequence at the 5' end of the cDNA from immunized rabbit 53 (Fig. 1). The cient (TCRct-/') mice (Fig. 2). We could inserts. Specific PCR products were identi- TCRP mAb stained proteins off and on the detect the pTaL-TCRP complex by precip- fied by Southern (DNA) hybridization with diagonal, corresponding to disulfide-linked itation with the mAb H57 to all another set of degenerate oligonucleotides

TCRP and -unlinked TCRP proteins, respectively, proteins (17) from lysates of TCRC-'- corresponding to the pTcx sequence between on July 9, 2019 in the lysates of both SCB.29 and thymocytes. Thus the pTa-TCRP com- the PCR primers. Hybridizing PCR frag- B6.2.16.BW cells (Fig. 1, top) whereas the plex does not represent a peculiarity of a ments were subcloned and sequenced. In this control GantiR-HRPO [goat anti-rabbit single cell line, but is also formed and way a fragment of approximately 180 bp was immunoglobulin (Ig) conjugated to horse- glycosylated in immature thymocytes. obtained that encoded part of the 5' un- radish peroxidase] produced no signal. The Cloning of the pTa chain complemen- translated region, the leader sequence, and pre-immune serum from rabbit 53 (PI53) tary DNA (cDNA). To obtain a pTa chain the sequence corresponding to the first five likewise produced no signal (except for cDNA sequence we prepared a cDNA li- amino acids of the pTao protein. This frag- some weak reactivity with the 30-kD mark- brary from the SCB.29 cell line in lambda ment was then used as a probe for conven- er on the diagonal) whereas the immune ZAP II (Stratagene). A lysate of the library tional screening of the SCB.29 cDNA li- serum (S53) stained in SCB.29 lysates a was prepared and amplified by PCR (poly- brary. Several independent lambda clones protein off the diagonal beneath the TCR,3 merase chain reaction) with degenerate or were isolated that contained the full-length protein that was not stained in B6.2.16BW nondegenerate oligonucleotides, correspond- coding sequence of pTcx. The sequence re- lysates (Fig. 1, middle). The staining of the ing to the most 3' part of the pTcx NH2- vealed an open reading frame of 618 nucle- pTac protein could be completely inhibited by the specific peptide, whereas the exten- sive background staining could not. Affin- Fig. 2. Identification of the pTa TCRa# ity purification on a peptide column finally chain in TCR-'- thymocytes (4 x SC'B.29 thymocytes yielded a reagent of exquisite specificity for 109 cells) and SCB.29 (2 x 107 C the pTax protein that became visible as a cells). Two-dimensional SDS- -200 200- streak of differentially glycosylated proteins PAGE of immunoprecipitates with -97 as well as a nonglycosylated dot (Fig. 1, H57 (17). The pTa chain is revealed -69 by CP53 serum. The spot on the -45 bottom). These results established that the diagonal may represent nondisul- 45- protein sequence was obtained from the in the precipitate. -30 pTa protein and that the antiserum could fide-linked pTa 30- Staining with mAb F23.1 was simi- '4 be used to analyze expression of the pTa- lar to that in Fig. 1. Methods were as -21 21- TCRP complex in various cells. in Fig. 1. Molecular sizes are indi- Because the pTot-TCRB complex in thy- cated in kilodaltons. SCIENCE * VOL. 266 * 18 NOVEMBER 1994 1209 otides encoding a hydrophobic leader se- Developmentally regulated expression fected cell line SCB.29 as well as in the quence of 23 amino acids. The extracellular of the pTot gene. The pTax RNA expres- nontransfected parental line SCI.ET.27F region comprised about 130 residues with sion was analyzed by Northern (RNA) (12), but not at all in the TCRao3-express- similarity to the constant (C) domain of the blotting: Total and polyadenylated ing hybridoma B6.2.16.BW (12). In mice, immunoglobulin supergene family with two [poly(A) 'I RNA was isolated from various there was some expression in the thymus, cysteines in position 31 and 91 correspond- tissues or cell lines and probed with pTa whereas lymph node (Fig. 4) and spleen ing to the invariant cysteines that form the cDNA (Fig. 4, left). The pTa message cells were negative. Expression of the pTa intrachain disulfide bond in the Ig domain. could be detected in total and poly(A)+ gene did not require rearrangement, as In addition there were two potential glyco- RNA from the immature, TCRI-trans- full-size RNA was abundant in the thymus sylation sites in this part of the molecule. A third cysteine in position 119 could be used for an interchain disulfide bond in the pTa- A TCRP complex. -23 1 V The extracellular Ig-like domain has MLLHEWAMARTWLLLLLGVRCQALPSGIAGTPFPSLAPPITLLVDGRQHMLVVCLVLDAAPPGLDNPVWFSAGNGSALDA only some weak (20 to 25 percent) homol- -SIGNAL PtElF= 0-CO ogy with the Ig-like domain of TCRox, Ig 58 V V light and Ig heavy chain constant domains, FTYGPSLAPDGTWTSLAQLSLPSEELEAWEPLVCHTRPGAGGQNRBTHPLQLSGESSTARSCFPEPLGGTQRQVLWLSLL as well as x5, being most homologous to the 138 constant domain of the IgA molecule. In RLLLFKLLLLDVLLTCSHLRLHVLAGQHLQPPPSRKSLPPTERIWTZ contrast to X5 (20), pTa does not contain a + I IM - -PKC- -PKC- J-like sequence but strictly ends at the Ig- Fig. 3. (A) Predicted pTat sequence. Amino acid B NH2 sequence of the open reading frame of pTa2 like C domain. Unlike X5, the pTa protein Downloaded from contains a transmembrane region and a cy- cDNA. The nucleotide sequence is available from the GenBank database, accession number ; toplasmic tail. The transmembrane region U1 6958. Amino acid numbering is given at the left. of about 20 hydrophobic residues includes Locations of the potential N-linked glycosylation two basic amino acids (arginine and lysine) sites (CHO), the predicted leader (peptide signal) NH2I- that are identical to the polar residues in and transmembrane (TM) sequences, and the the TCRoQ chain kinase C sites

of http://science.sciencemag.org/ the transmembrane part potential protein phosphorylation S S and are separated by the same number of (PKC) are shown. The sequences PPSRK (169) pIa hydrophobic residues (18). In the TCRa and PPTHR (176) are similar to the PPGHR motif pTa s S chain, these polar residues are essential for found in CD2. The three cysteines in positions 31, the assembly and transport of the TCR43- 91, and 1 19 are marked (V). DNA sequence anal- 7-S-S ysis was done with the GCG program (Genetics CD3 complex (18). The cytoplasmic region computer group, program manual for the GCG of about 31 residues is rich in proline resi- package, version 7, April 1991). Homology 3 S U R+ dues and could constitute a SH3-domain searches in the EMBL (release 33.0, March 1994), ( /(1(1( binding region. In addition, there are GenBank (release 83.0, February 1994) and UU+U PPSRK and PPTHR sequences similar to SWISSPROT (release 28.0, February 1994) data 0000 the PPGHR motif present in the cytoplas- banks were done with the FASTA (39) program. COOH mic tail of CD2, known to be involved in Sequence comparisons and alignments with in- on July 9, 2019 CD2-dependent T cell activation (19) and munoglobulin superfamily domains were done COOH containing two potential phosphorylation with the PILEUP (40) program. Single-letter abbreviations for the amino acid residues are: A, Ala; C, Cys; sites for protein kinase C 3). D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, lie; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gin; R, Arg; S, Ser; (Fig. T, Thr; V, Val; W, Trp; and Y, Tyr. (B) The pTa-TCR,3 complex. _, Corresponds to the N-linked The sequence analysis indicates that the glycosylation sites. P, represents protein kinase C phosphorylation sites in the pTa cytoplasmic tail. Two pTa protein is well suited for pairing with basic amino acids, arginine (R') and lysine (K'), are located in the pTa transmembrane region. the TCR, chain. Even though there is only moderate identity to any particular member of the Ig supergene family, this protein bears Fig. 4. pTot mRNA expression in the hallmarks of this family. The cysteine in various cells and tissues. Total or as transmembrane poly(A)+ RNA was isolated from cell position 119 as well the lines and tissues and the RNA was part seem designed to form a complex with separated on agarose gels. After TCRI and CD3 proteins. The pTa-TCRB transfer, the RNA was probed with heterodimer has an asymmetrical shape be- a labeled cDNA specific for pTcx; af- cause the TCRP has two and the pTcx only ter the original probe was stripped one extracellular Ig-like domain. One might from the filter, it was again hybrid- expect therefore that another short protein ized with a labeled ,-actin probe may be noncovalently attached to the mol- (41). ecule, perhaps like the VpreB domain (20) that is supposed to bind to the pre-B cell receptor. The relatively long cytoplasmic tail of pTa is different from the short cytoplasmic tail of the TCR a and P chains and the Ig heavy chains. Because of the potential SH3-domain binding sequence, phosphorylation sites, and the CD2 like sequences, the pTa chain may be directly involved in signal transduction by the pre-TCR. 1210 SCIENCE * VOL. 266 * 18 NOVEMBER 1994 B .l I1mW1BX WAV of RAG-2-'- (recombination activating Thus pTa is expressed in the tlhymus be- chains. When these cells were supertrans- gene 2) rearrangement-defective mice fore TCRI and TCRoL genes, but there fected with the pTot construct, intracellu- (21) and it was expressed higher in RAG- may be some overlap in the exipression of lar pTa-TCRP dimers were observed, but 2-/- thymocytes than in thymocytes from pTot and TCRa genes. Coexp)ression of in contrast to dimers in the immature T normal mice (Fig. 4). The latter result pTcx-TCRB and TCRot-TCR[3 heterodimers cell line SCB.29, the disulfide-linked pTa suggested that the pTco gene was predom- on the cell surface of TCRm-itransfected proteins in the mature T cell line were inantly expressed in the more immature SCB.29 cells has been observed (2?2). There- poorly or not at all glycosylated, indicat- thymocytes. fore this coexpression is theoretic-ally possi- ing that most of the heterodimer did not Expression in early thymocytes was ble in immature thymocytes. leave the endoplasmic reticulum (Fig. 6). confirmed by PCR analysis with oligonu- The transfection of pTa in a TCRa- Also, surface expression of TCRI in the cleotides specific for the pTcx gene (Fig. deficient, mature T cell line. 1[he above pTa transfectants was not increased, 5). The developmentally regulated pTot results show that a glycosylated pTcx-TCRP whereas TCRca4 surface expression was message was expressed in different complex is transported to the cell surface in observed after TCRQ. transfection. This amounts at successive stages of thymic immature T cells. Because TCRI13 has two result was consistent with the hypothesis development. There was high expression and pTcx only one extracellular ]Ig-like do- that another developmentally regulated in CD44-CD25+ and CD44-CD25- main, one would expect that the pTa- protein (VpreT) may be required for prop- double negative (CD4-CD8-) cells but TCRP complex has an addition;al, nonco- er assembly, glycosylation, and transport there was weaker expression in earlier and valently associated Ig-like domcain, much to the cell surface of the pTao-TCR13 com- later stages of development. Small single like the VpreB domain in the 1pre-B cell plex. The result is of course equally con- positive (expressing only CD4 or CD8) receptor (23). One might furthier expect sistent with the hypothesis that mature T thymocytes had no detectable pTat RNA. that without such an additional p)rotein the cells may contain proteins that prevent The difference in the intensity of the pTcx pTa-TCR3 complex may not be transport- pTa-TCRP complexes from leaving the Downloaded from RNA bands between RAG-2-/- and nor- ed to the cell surface. endoplasmic reticulum and that the pTa- mal thymocytes in Fig. 4 is explained by To test whether or not the p'Tra-TCRP TCRB complex can only be transported to the fact that most cells in the normal complex was sufficient for cell s;urface ex- the cell surface in the absence of these thymus are small CD4+CD8+ cells, pression, we transfected a pTot construct retention molecules. whereas in the RAG-2-'- thymus most into the 58ot-py.1 mature T cel11 line (8), Regulation of T cell development by cells are CD44-CD25+ and because which expresses a TCRP gene and CD3 the pre-TCR complex. In conjunction with http://science.sciencemag.org/ CD44- CD25 + cells express more pTao proteins that are essential for siurface ex- experiments in TCR13 transgenic, rear- RNA than small CD4+CD8+ thymocytes. pression of the o43 TCR, but laccks TCRcx rangement-defective SCID mice (6-8) and analogous experiments in TCRP transgenic RAG-/- mice (21, 24), our results are con- RAG2-1- C57BU6 Source sistent with the hypothesis that the pre- 1 11~~~~~~~~~~~~I TCR complex is sufficient to promote T cell development in the absence of other D TCR chains encoded by rearranging genes. (D 0 U) uO v) 0) C14 C14 C4 E Phenotype Although the experiments in the TCR,3 0 CD 0 0 C) c transgenic, rearrangement-deficient mice to E on July 9, 2019 ,c 0 N 8 0jC) 00 have indicated that TCR13 proteins are suf- I* 0D 0* 0. ficient to induce T cell C-) c) I0 0 u,)n- C,) maturation, studies pTa in TCR-/- mice suggest that a protein is not essential for the expressionTCRPof I-E il l CD4 and CD8 molecules in development l~ because CD4+CD8+ cells are present in HRPT TCRP-/- mice (24). The possibility was considered that in 100.0 88.4 99.9 99.0 89.0 99.0 100.0 100.0 Purity (°) CD4+CD8+ thymocytes PuRity2(%) TCR,13 mice were of the y8 lineage (be- Fig. 5. Semiquantitative analysis of pTa expression on lymphocyte subsets. Thymocytes frorn cause they were absent in TCRB-/- and normal C57BU6 mice were separated by fluorescence-activated cell sorting (FACS) or ccomplement-npRAG-2-'- TCRy-/- mice) and that the TCRB protein Subsets of T cells were to chain reaction as descr dependent killing. subjected polymer (PCR) was essential for of CD4 or CD8 Samples were stained with ethidium bromide. SP, single positive (CD4 or CD8) T cells; HPR- , ihypoxan- expression coreceptors in the ao3 lineage (24). There thine-guanine phosphoribosyl transferase. are, however, no experiments supporting this view. In contrast, reconstitution of thy- Fig. 6. Transfection of a pTa construct into a Cl) muses from SCID (25, 26) or RAG-/- mice mature, TCRa-deficient T cell line. Immunopre- CM. (22) with either pro-T cells or ^y8T cells cipitations of TCRI from SCB.29, SCI.ET.27F (the LL have shown that thymocytes devoid of TCR cell the transfec- r- nontransfected SCID line), pTa V11 Fi U)% j 0 proteins can express CD4 and CD8 mole- tants of 58a- ,y. 1 (a TCRa-mature T cell line) (43) cules as long as other TCR-bearing cells are named 58a- y.1 -33 and represented by clones 0 0 U.)U) U)0 U)CU) >>0 present. Thus the pre-TCR complex may 7, 9, and 10, and 58a- y.1 cells were separated not directly regulate CD4 and CD8 expres- under nonreducing (NR) and reducing (R) condi- 200- tions and immunoblotted with the CP33 pTa an- 97- sion through intracellular signaling. tiserum. 69- Intracellular signaling by the pre-TCR NR 45- R complex may, however, be essential to ob- 30- ___ tain large numbers of immature thymocytes. 21 -_ This can be deduced from experiments that showed that the proportion of productive SCIENCE * VOL. 266 * 18 NOVEMBER 1994 1211 m rearrangements in CD44-CD25- (28) and gawa, J. G. Seidman, Science 241, 1089 (1988). recommendations (Du Pont). Filters were first hybrid- 11. H. von Boehmer et al., Proc. Natl. Acad. Sci. U.S.A. ized at 650C for 8 hours and hybridized with the same total thymocytes (29) from TCRa-/' mice 85, 9729 (1988). solution [1 mM EDTA, 0.5 M NaH2PO4, 7 percent were much higher than expected if cells of 12. M. Groettrup et al., Cell 75, 283 (1993). SDS, and salmon sperm DNA (100 pg/mI)] containing this phenotype could similarly expand, irre- 13. M. Groettrup and H. von Boehmer, Eur. J. Immunol. the specific probe at 650C overnight. To remove any 23,1393(1993). unspecifically bound probe, the membranes were spective of whether or not they carried 14. H. von Boehmer, Cell 76, 219 (1994). washed first with 1 mM EDTA, 40 mM NaH2PO4, 1 productive TCRPi genes. Thus the pre-TCR 15. M. Groettrup and H. von Boehmer, Immunol. Today percent SDS at 650C. Filters were hybridized first with complex may regulate survival and expan- 14, 6101 (1993). a cDNA probe corresponding to the entire coding 16. U. D. Staerz, H.-G. Rammensee, J. D. Benedetto, M. sequence of pTcx and subsequently, after dishybrid- sion of immature T cells rather than regu- J. Bevan, J. Immunol. 134, 3994 (1985). ization, with a probe specific for ]3-actin message. lating directly CD4 and CD8 gene expres- 17. R. T. Kubo, W. Born, J. Kappler, P. Marrack, M. J. Washed membranes were exposed to Kodak sion in the lineage of a13 T cells. Pigeon, ibid. 142, 2736 (1989). X-OMAT AR film at -700C for 1 to 5 days. The role of the pre-TCR complex in T 18. P. Cosson, S. P. Lankford, J. S. Bonifacio, R. D. 42. CD44+CD25-, CD44+CD25+, and CD44-CD25+ Klausner, Nature 351, 414 (1991). double-negative (CD4-CD8-) thymocytes were ob- cell development is supported by other 19. H. C. Chang, J. Exp. Med. 172, 351 (1990). tained from RAG-2-'- mice. CD44-CD25- thymo- studies. In SCID (30) as well as in TCRP- 20. H. Karasuyama, A. Kudo, F. Melchers, ibid., p. 969. cytes were obtained from C57BL/6 mice by comple- or RAG-deficient (31, 32) mice, the ef- 21. Y. Shinkai et al., Science 259, 822 (1993). ment-dependent killing of thymocytes with antibod- 22. M. Groettrup, thesis, University of Basel, Switzerland ies to CD4 and CD8. The thymocytes were then fects of productive TCRP transgenes on (1993). stained with antibodies specific for CD4, CD8, thymocyte development can be mimicked 23. A. Kudo and F. Melchers, EMBO J. 6, 2267 (1987). CD44, CD25, and CD3 surface markers to obtain the by treating thymocytes of these mice with 24. P. Mombaerts et al., Nature 360, 225 (1992). desired populations by cell sorting. The purity of the 25. E. W. Shores, S. 0. Sharrow, A. Singer, Eur. J. Im- subsets was deduced from reanalysis of sorted cells. antibodies to CD3e. These results support munol. 21, 973 (1991). In order to quantitate pTa RNA, cells (5 x 104) were the hypothesis that the TCRP protein 26. F. Lynch and E. M. Shevach, Int. Immunol. 5, 991 directly sorted in 500 ,u of RNAzol (TM Cinna Scien- exerts its function through a pre-TCR (1993). tific), and total RNA was extracted according to the 27. H. R. Rodewald, unpublished data. manufacturer's protocol. The cDNA was prepared complex that associates with CD3 pro- 28. E. C. Dudley, H. T. Petne, L. M. Shah, M. J. Owen, A. with random hexamer primers and reverse-tran- teins. It is not clear from these studies C. Hayday, Immunity 1, 83 (1994). scribed with Superscript kit (Gibco BRL). Dilutions Downloaded from whether signaling through CD3 is all that 29. C. A. Mallick, E. C. Dudley, J. L. Viney, M. J. Owen, (1:3) of cDNA in water were then used in PCR A. C. Hayday, Cell 73, 513 (1993). amplification reactions. Primers used were oligonu- is required or whether the cytoplasmic 30. C. N. Levelt, A. Ehrfeld, K. Eichmann, J. Exp. Med. cleotides recognizing sequences in the 5' and 3' part of the pTot chain plays an essential 177, 707 (1993). regions of the pTa (5'-CTGCAACTGGGTCAT- role. A potential role of the cytoplasmic 31. C. N. Levelt, P. Mombaerts, A. Iglesias, S. Tonegawa, G0TTC-3' and 5' TCAGA0GGGTGGGTAAGATC- to K. Eichmann, Proc. Natl. Acad. Sci. U.S.A. 90,11401 3') and HRPT (5'-CACAGGACTAGAACACCTGC-3' tail of pTa needs be analyzed under (1993). and 5'-GCTGGTGAAAAGGACCTCT-3') genes. Am-

more physiological conditions. 32. H. Jacobs et al., Eur. J. Immunol. 24, 934 (1994). plification was done for 35 cycles at an annealing http://science.sciencemag.org/ A role of p56lk in the signaling by the 33. T. J. Molina et al., Nature 357, 161 (1992). temperature of 550C with a thermal cycling machine pre-TCR is implicated by studies in mice 34. S. D. Levin, S. J. Anderson, K. A. Forbush, R. M. (Perkin-Elmer Cetus). A 1 5-,ul portion of each ampli- Perlmutter, EMBO J. 12, 1671 (1993). fied product was separated through a 1.2% agarose that are Lck-defective (33) or carry a dom- 35. P. Mombaerts, S. J. Anderson, R. M. Perlmutter, T. gel by electrophoresis and stained with ethidium inant negative Lck mutation (34). The W. Mak, S. Tonegawa, Immunity 1, 261 (1994). bromide. Lck dominant negative mutation results in 36. S. J. Anderson, K. M. Abraham, T. Nakayama, A. 43. The 58a- ,y. 1 T cell line was obtained by transfection Singer, R. M. Perlmutter, EMBO J. 11, 4877 (1992). ofthe C58 TCRat-TCR,- T cell line with aTCRP gene a developmental arrest that resembles that 37. H. Yu and L. S. Schreiber, Structural Biol. 1, 417 (8). This cell line was supertransfected with a pTa of rearrangement-deficient mice. On the (1994). construct. To this end, full-length cDNA of pTa sub- other hand the introduction of an active 38. Rabbit 53 was immunized with a peptide corre- cloned in Bluescript SK+ (clone D2.1) was amplified sponding to the NH2-terminal residues of the pTot with primers 5'-TAGCCTCGAGCTGCMCTGGGT- lck gene in rearrangement-deficient mice shown in Fig. 3. (The peptide contains an Ala instead CATGCTTC-3' and 5'-CAGAGAATTCTCAGACG- (35) or in normal mice (36) causes the of a Ser in position 12.) Either SCB.29 or B6.2.1 6BW GGTGGGTAAGATC-3' that anneal in the 5' and 3' same effects as a transgene. Thus cells were lysed in Triton lysis buffer and the lysate of untranslated region of pTa just upstream of the ATG on July 9, 2019 TCRP 107 cells was immunoprecipitated with mAb F23.1 start and downstream of the TGA stop codon, re- p561ck may be part of the signaling cascade (16) coupled to agarose beads. The precipitates spectively, and introduced specific artificial cloning of the pre-TCR complex. Because the cy- were separated by SDS-1 0 percent polyacrylamide sites: Xho in the former case, Eco RI in the latter. The toplasmic tail of the pTa chain contains a gel electrophoresis (PAGE), first under nonreducing resulting PCR fragment of about 660 bp was purified, (NR) and then under reducing (R) conditions. After digested with Xho and Eco RI, and subcloned into proline-rich, potential SH3-binding do- equilibration in transfer buffer, the proteins were the corresponding polylinker restriction sites of Blue- main (37), p561Ck may be recruited by the transferred by wet blotting onto polyvinyliden difluo- script SK+. The pTa sequence between the Nco site pTaL protein into the pre-TCR complex. ride membranes. Filters were incubated with either at the second in-frame ATG and the one Bgl II site The and expression analysis of the preimmune serum (P153, at a dilution of 1 100), downstream of the stop codons was replaced by the cloning the immune serum (S53, at 1:100) or the affinity corresponding sequence from D2.1 in order to elimi- the pTa gene has established that the pre- column-purified (CP53, at 1: 100) in phosphate-buff- nate potential mutations introduced during the PCR TCR is expressed in immature thymocytes ered saline containing 0.4 percent Tween and 2 per- amplification. The resulting plasmid was digested with and opens the way to determine the precise cent fetal calf serum. For peptide competition, pep- Xho and Not I to release the cDNA fragment encod- tide (35 pgg/ml) was first incubated with S53 serum or ing pTa, which was then inserted in the correspond- structure and function of the pre-TCR com- mAb F23. 1 for 1 hour. For detection of TCRB protein ing sites of expression vector BCMGSHyg (20). The plex in T cell development. a F23.1 -biotin conjugate was used. After the first- construct was transfected into 58a-3y.1 cells by stage reagents were removed, GantiR-HRPO at electroporation. AND NOTES 1: 20,000 or streptavidine-HRPO (1: 5000) was add- 44. We thank D. Avila, for microsequencing of the pTa REFERENCES ed for visualization of the rabbit and F23.1 antibod- protein; J. Kaufman, M. Reth, and J. Cebrian for ies, respectively. help with the analysis of the pTa sequences; 1. D. H. Raulet, R. D. Garman, H. Saito, S. Tonegawa, 39. W. M. R. Pearson and D. J. Lipman, Proc. Natl. C. Benoist, D. Mathis, and M. Wiles for advice on Nature 314, 101 (1985). Acad. Sci. U.S.A. 85, 2444 (1988). pTa cloning; J. Garcia-Sanz for advice on the RNA 2. R. H. Snodgrass, Z. Dembic, M. Steinmetz, H. von 40. D. F. Feng and R. S. Doolittle, J. Mol. Evol. 35, 359 analysis; S. Goude, V. Stauffer, and C. Geiger for Boehmer, ibid. 315, 232 (1985). (1987). secretarial help; and K. Hafen and C. Laplace for 3. M. Bosma and A. M. Carrol, Annu. Rev. Immunol. 9, 41. Total RNA was prepared from cell suspensions (thy- expert technical assistance. The Basel Institute for 323 (1991). mocytes from C57BU6 mice and RAG-2-'- mice, Immunology is supported by Hoffmann-La Roche, 4. Y. Shinkai, Cell 68, 855 (1992). and lymph node cells from C57BU6 mice) and from Basel. Supported in part by the Institut National de 5. P. Mombaerts et al., ibid., p. 869. cell lines (SCB.29, SCI.ET.27F and B6.21 6.BW cells) la Sante and Recherche Medicale, Paris, and by 6. H. von Boehmer, Annu. Rev. Immunol. 8,531 (1990). with RNAzol B (TM Cinna Scientific, Friedswood, the Faculte Necker Enfants Malades, Descartes 7. H. Kishi etal., EMBO J. 10, 93 (1991). TX). The mRNA was prepared from total RNA with Universite, Paris. H.v.B. is affiliated with the Faculte 8. M. Groettrup, A. Baron, G. Griffiths, R. Palacios, H. the use of the Oligotex TM mRNA Kit (QIAGEN). Both Necker, Paris, and also supported by the Human von Boehmer, ibid. 11, 2735 (1992). total RNA and mRNA were separated on 1.2 percent Frontier Science Project Organization (HFSPO). 9. Y. Uematsu et al., Cell 52, 831 (1988). agarose-formaldehyde gels and transferred to Gene- 10. R. G. Fenton, P. Marrack, J. W. Kappler, 0. Kana- Screen membranes according to the manufacturer's 9 September 1994; accepted 18 October 1994

1212 SCIENCE * VOL. 266 * 18 NOVEMBER 1994 Analysis and expression of a cloned pre-T cell receptor gene C Saint-Ruf, K Ungewiss, M Groettrup, L Bruno, HJ Fehling and H von Boehmer

Science 266 (5188), 1208-1212. DOI: 10.1126/science.7973703

ARTICLE TOOLS http://science.sciencemag.org/content/266/5188/1208 Downloaded from

REFERENCES This article cites 36 articles, 9 of which you can access for free http://science.sciencemag.org/content/266/5188/1208#BIBL http://science.sciencemag.org/ PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

on July 9, 2019

Use of this article is subject to the Terms of Service

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. The title Science is a registered trademark of AAAS.