Signal transduction through CD40
Eur. J. Immunol. 1990. 20: 1747-1753
Seiji Inuiv, Tsuneyasu Kaisho, Hitoshi Kikutani, Ivan StamenkovicAo, Brian SeedAn, Edward A. Clarko and Tadamitsu Kishimoto Division of Immunology, Institute for Molecular and Cellular Biology, Osaka University, Osaka, Department of Geneticso, Harvanl Medical School and Department of Molecular BiologyA, Massachusetts General Hospital, Boston and Department of Microbiologyo, University of Washington, Seattle
1747
Identification of the intracytoplasmic region essential for signal transduction through a B cell activation molecule, CDN* CD40 is a 45-kDa glycoprotein expressed on human B lineage cells. Anti-CD40 induces the proliferation of B cells and the extracellular region of CD40 is related to those of a certain kind of growth factor receptors. Therefore, it has been proposed that CD40 might be a receptor for a molecule involved in the growth regulation of B cells.The cDNA coding for CD40 was transfected into the murine B lymphoma cell line M12 and the murine thymoma cell line EL4.The growth of both M12 and EL4 transfectants expressing human CD40 was inhibited by anti-CD40. Phorbol 12-myristate 13-acetate (PMA) augmented the growth inhibitory effects of anti-CD40 on transfectants. The CD40 molecule was constitutively phosphorylated not only in human tonsil B cells but also in transfectants expressing CD40. PMA augmented the phosphorylation of CD40 in these cells. These results indicate that in spite of the growth inhibitory effect of anti-CD40, the augmentative effect of PMA is conserved in CD40+ transfectants and suggest that the transfectant might be useful for the study of signal transduction mechanism through CD40. To investigate which part of the CD40 molecule is important for signal transduction, transfectants expressing mutant CD40 cDNA were established and their growth response to anti-CD40 was evaluated.The mutant molecule, which had an Ala for Thr substitution at position 234, and the deletion mutants lacking Thr234were inactive in growth signal transduction, indicating that Thr234itself or the region around Thr234is essential for signal transduction through CD40.
1 Introduction A number of lymphokines or cytokines involved in the regulation of B cell growth or differentiation have been identified, molecularly cloned and shown to have a pleiotropic effect on various cells including T cells, MWmonocytes as well as on B cells [l-41. The receptors for these molecules are expressed not only on B cells but also on other lineage cells. On the other hand, there certainly exist some lineage-restricted surface molecules which are involved in cell activation, growth or differentiation. Tcells express the CD3 complex and the CD2 molecule.The CD3 complex is associated with the TcR and transduces the signals necessary for Tcell activation [5]. Certain anti-CD2 antibodies were shown to induce the proliferation of Tcells [6-81. Moreover, the interaction of CD2 with another cell surface molecule, the lymphocyte function-associated antigen 3 (LFA-3), was found to play an important role in antigen-independent cell-to-cell communication [9, 101. Certain B lineage-restricted membrane molecules such as CD19, CD20 or CD40 have also been proposed to be
involved in B cell activation or growth [ll-171, although their physiological roles remain to be resolved. CD40 is a 45-kDa glycoprotein specifically expressed on the surface of B lineage cells and dendritic cells of hematopoietic cell origin [15, 181. CD40 was also shown to be expressed on tumor cells, mostly of epithelial origin [16, 19-21]. Expression of CD40 was enhanced when B cells were activated with PMA or anti-IgM [16]. The proliferation of normal B cells could be induced by anti-CD40 when they were preactivated with anti-IgM, PMA, or anti-CD20 [15]. These results indicate that CD40 is one of growth factor receptors involved both in tumor growth and in normal B cell proliferation.
The CD40 cDNA was cloned by Stamenkovic et al. [17] using a high-efficiency expression system in mammalian cells. These sequence analysis revealed that CD40 has 20 cysteine residues in the extramembrane region while only one in the cytoplasmic region. The receptors of a certain kind of growth factors such as epidermal growth factor (EGF), CSF-1, and insulin also have cysteine-rich extramembrane regions. In fact, the positions of the cysteine residues are conserved in the extramembrane regions of [I 84471 CD40 and the nerve growth factor (NGF) receptor [22-251. * This study was supported in part by Grant-in-Aid for Specially However, no tyrosine kinase domain, which is considered Promoted Research from the Ministry of Education, Science crucial for signal transduction in a certain family of and Culture and by National Institutes of Health Grant receptors [26], was observed in the cytoplasmic region of GM37905. CD40 or NGF receptor [17, 251. Thus CD40 appears to Present address: Basel Institute for Immunology, CH-4005 belong to one of the families of growth factor receptors that Basel, Switzerland use signal transduction mechanisms differing from those Correspondence: Tadamitsu Kishimoto, Institute for Molecular with a tyrosine kinase domain.
and Cellular Biology, Osaka University, 1-3-Yamada-oka, Osaka 565, Japan Abbreviations: growth factor
EGF Epidermal growth factor NGF Nerve
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
To elucidate the function of the signal transduction mechanisms through CD40, we transfected CD40 cDNA into murine cell lines. Anti-CD40 and PMA inhibited the proliferation of transfectants expressing CD40 and PMA 0014-2980/90/0808-1747$3.50+ .25/0
1748
Eur. J. Immunol. 1990.20: 1747-1753
S. Inui,T. Kaisho, H. Kikutani et al.
enhanced the phosphorylation of CD40 in transfectants as well as in normal human B cells. These results suggest that the CD40 molecule on transfectants can transduce the growth signal and that the transfectants might be useful for the analysis of signal transduction mechanisms. Using this transfectant system and mutant CD40 cDNA, we further showed that the cytoplasmic region around Thr234was essential for signal transduction through CD40.
2 Materials and methods 2.1 DNA The CD40 cDNA was cloned by Stamenkovic et al. [17] with an expression vector (CDM8) carrying the transcription unit composed of the human cytomegalovirus enhancer and the HIV long terminal repeat. The plasmid pSV2neo carries the neomycin resistance gene and imparted resistance to G418 [27]. Twenty microgram of wild type or mutant CD40 cDNA and 2 pg of pSV2neo were linearized by Sac I1 and Barn HI, respectively, and employed for transfection of 1 x 107cells.
5’-GAGATGCGACCCTCCC-3r(Ser2494 Gly249), 5’-CCTGCACTGCGATGCGACTC-3‘(Ser252 4 AlaZ2), 5 ’-CCGGTTGGCCTCCATGTAAA-~’(CYS~ + Gly’*), 5’- GTA A AGTCTACTGCACTGG A-3' (G ~ +uSTOP), ~ ~ and yielded the mutant cDNA 203A, 227A, 234A, 225A, 249G, 252A, 2386 and Del (232), respectively. The plasmids containing mutant CD40 cDNA were isolated and sequenced around the modified region by the dideoxy sequencing method [31]. The double-stranded DNA were prepared from E. coli infected with each M13mp19 mutant clone and digested with Hind I11 and Sma I. The resulting DNA fragments were subcloned into CDM8 which had been previously prepared. Another deletion mutant Del (201) (Pro202 + STOP) was made as follows. The CD40 cDNA was digested with Xho I and Bal I, and ligated to the linkers, 5’-CCAGAAGTGACTGCA-3’ and 5’-GTCACTTCTTGG-3‘, to change Pro202into a termination codon, and provide a Bal I restriction recognition site. The resulting fragment was subcloned into the Pst I and Xho I sites of CDM8. 2.4 Cell staining and sorting
2.2 Transfection of DNA DNA was transfected into the murine B lymphoma cell line M12 [28] and the thymoma cell line EL4 [29] by the electroporation method [30]. Ten million exponentially growing cells were harvested, washed twice with PBS, and then resuspended at a concentration of 1 x lo7 cells/ml in ice-cold PBS (120.7 mM KC1, 30.8 mM NaC1, 8.1 mM Na2HP04, 1.46 mM KH2P04) including 5 mM MgC12. Plasmid DNA (see above) were added to the cell suspension and kept on ice for at least 15 min. An electric pulse of 1.5 kV was applied by a Promega X cell-2000 power supply (Seikagaku Kogyo, Co., Ltd.,Tokyo, Japan).The cells were kept on ice for 7 min, transfered to the RPMI 1640 medium supplemented with 10% FCS and kept at room temperature for 10 min. After two days of culture in normal medium, M12 and EL4 transfectants were selected in 1.0 and 0.5 mg/ml G418 (Gibco Laboratories, Grand Island, NY), respectively.
One million cells were incubated with biotinylated G28-5 (anti-CD40; [15]) for 20 rnin at 4 ° C washed three times with staining buffer (PBS, 2% FCS, 0.02% NaN3), and counterstained with FITC-conjugated avidin. Propidium iodide was included for the last 5 min to gate out the data provided by dead cells. Stained cells were analyzed and sorted on a fluorescence activated cell sorter (FACS 440, Becton Dickinson, Mountain View, CA). 2.5 Proliferation assay Cells were cultured at 1 x 104/wellin 96-well microtiter plates containing 200 p1 of RPMI 1640 medium supplemented with 2% FCS, antibiotics, L-glutamine and 2-ME with or without various stimulations. After 3 days, cells were pulsed with 0.5 pCi = 18.5 kBq of [3H]thymidine/well for the last 3 h. Cells were then harvested with a cell harvester onto glass-fiber filters, and the radioactivity was measured.
2.3 Construction of mutant CD40 cDNA 2.6 The phosphorylation of CD40 To construct mutated CD40 molecules, the plasmid CDM8 carrying the CD40 cDNA was digested with restriction enzyme Not I, blunted with Klenow fragment of E. coli DNA polymerase I and further digested with Hind 111.The CD40 cDNA blunt-Hind 111 fragment was inserted into M13mp19 previously digested with Hind 111and Hinc I1 and employed to produce a single-stranded DNA template for oligonucleotide-directed mutagenesis. Oligonucleotidedirected mutagenesis was performed using an in vitro mutagenesis kit (Amersham Int., Amersham, GB). All oligonucleotides used were synthesized on 381A DNA synthesizer (Applied Biosystems, Fostercity, CA). Singlebase substitutions were introduced using appropriate mismatched 20-mer oligonucleotides as follows: 5’-CCTTAlTGGCTGGCITCTTG-3’(Thr203 + Ala203)),
Ten million actively growing cells were washed three times by phosphate-free RPMI 1640 medium and incubated at 1 x 106/ml for 2 h in phosphate-free RPMI 1640 medium supplemented with 10% FCS. For E rosette-negative cells, 1 x los cells were used and incubated at 1 x 107/ml.Cells were labeled with 1mCi [”2P]orthophosphate in 2 ml for 2 h at 37°C. For the last 30 min, PMA was added at 2 ng/ml. Cells were lysed with 0.4 ml of a solution containing 50 mM Hepes, pH 7.4, 1% Triton X-100,lO mM sodium pyrophosphate, 100 mM NaF, 4 mM EDTA, 2 mM sodium orthovanadate, 1 mM PMSF and 100 U/ml aprotinin. Cell lysates were clarified by centrifuging at 14000 x g for 60 rnin at 4°C. SN were precleared with an irrevalent antibody, 3-5 [30], and protein G-Sepharose (Pharmacia, Piscataway, NJ). Immunoprecipitation was carried out by G28-5 and 5‘-GAGCAGCAGCG?TGGAGCCA-3’(Thr227 + 5r-CATGTAAAGCCTCCTGCACT-3f(Thr234 + Ala234), protein G-Sepharose. Immunoprecipitates were washed 5’-AGCAGTGTTGGCGCCAGGAA-3’(Ser225 + Ala2=), four times with a washing solution containig 50 mM Hepes,
~
Signal transduction through CD40
Eur. J. Immunol. 1990. 20: 1747-1753
1749
pH 7.4, 1% Triton X-100, 0.1% SDS, 150 mM NaCl, 100 mM NaF and 2 mM sodium orthovanadate, and eluted by boiling in electrophoresis sample buffer (2% SDS, 10% glycerol, 5% 2-ME and 62.5 mM Tris-HCI, pH 6.8). Samples were analyzed by SDS-PAGE, and subjected to autoradiography.
3 Results 3.1 Establishment of murine cell lines expressing CD40
The murine B lymphoma cell line M12 was transfected with the CD40 cDNA and pSV2neo. After two weeks of G418 selection, drug-resistant clones were obtained and screened by indirect immunofluorescence as described in Sect. 2.4. After several cycles of sorting, the expression level of CD40 became stable without further sorting. FCM analysis pattern of a representative CD40+ clone, K2C3, is shown in Fig. 1A. This clone was employed in the following functional analysis. Similarly, the murine thymoma cell line EL4 was transfected with CD40 cDNA and CD40+ clone 2B1 was established (Fig. 1B). 3.2 Functional analysis of CD40- expressing transfectants Anti-CD40 was shown to induce the proliferation of normal B cells in the presence of co-stimulants such as anti-IgM, PMA, or anti-CD20 [15]. K2C3 was stimulated by various concentrations of anti-CD40 in the absence or presence of PMA to analyze whether the CD40 expressed on the murine cells could transduce a growth signal (Fig. 2). In contrast to normal human B cells, anti-CD40 inhibited the proliferation of K2C3. The growth inhibitory effect of anti-CD40 was dose dependent and observed at doses as low as 12.5 ng/ml. Anti-CD40 inhibited the growth of K2C3
(BI
0.1
1
10
100 0.1 Fluorescein (log)
1
10
100
Figure 1. FCM analysis of CD40-expressing transfectants. Parent cell lines (M12 and EL4) and transfectants (CD40-expressing M12, K2C3 and CD40-expressing ELI, 2B1) were stained with biotinylated G28-5 (anti-CD40) and FITC-conjugated avidin. (A) M12(-) and K2C3(-); (B) ELI(-) and 2B1(-). Ten thousand cells were analyzed per sample on the FACS 440. Histograms show relative cell number (y-axis) vs. log fluorescence (x-axis). Dotted line(---) is a fluorescence profile of the transfectants stained with FITCconjugated avidin alone.
b
io 160 Antibody concentration
ng/ml
Figure 2. The growth inhibitory effect of various concentrations of anti-CD40 on the murine B cell line M12 expressing CD40. Ten thousand cells were cultured for 3 days with various concentrations of anti-CD40 in the presence or absence of PMA. Proliferation was measured as the incorporation of [3H]dThd during the last 3 h. MlZ(0); MI2 with 2 ng/ml PMA(0); K2C3(0); K2C3 with 2 nglml PMA(W). Results were expressed as the percentage of [3H]dThd uptake of cells with no stimulation. All experiments were carried out in triplicate cultures.
up to 90% in the presence of PMA. The proliferation of other CD40+ transfectant clones such as K2C4, which expresses lower levels of CD40 than K2C3, was also inhibited similarly by anti-CD40 and PMA (data not shown). However, antLCD40 or PMA had very little effect on the parental M12 cells even when added together. The antibody concentrations required for growth inhibition of K2C3 were similar to those required for augmenting the proliferation of normal human B cells [15, 161. Although PMA alone also did inhibit the growth of transfectants, such an inhibitory effect is variable among transfectants (data not shown). In spite of variable effects by PMA alone, PMA consistently augmented the growth inhibition by anti-CD40. To test the specificity of anti-CD40-induced growth inhibition, we examined the effect of control antibodies, either antLCD23 (3-5; [32]) which was of the same isotype (IgGI) as anti-CD40, or anti-I-E (13/4) which reacted with both transfectants and parental cells. Control antibodies at 250 ng/ml showed no further inhibitory effect on the proliferation of PMA-stimulated transfectants (Fig. 3A). These results demonstrate that the growth inhibition of K2C3 is CD40 specific. To study whether the growth inhibitory signal is B cell specific, the effect of anti-CD40 on CD40+ murine T cell transfectant was analyzed. Likewise the proliferation of 2B1 derived from EL4 was inhibited by antLCD40 and PMA (Fig. 3B). Neither antLCD23 nor anti-Thy-1.2 inhibited the proliferation of EL4 and 2B1 (data not shown).These results indicate that the growth of transfectants are negatively regulated by the signal through CD40 in both murine B and Tcell lines.
1750
S. Inui,T. Kaisho, H. Kikutani et al.
Eur. J. Immunol. 1990. 20: 1747-1753
Figure 4 . CD40 molecules expressed on transfectants as well as normal human B cells are phosphorylated. Cells were labeled with 32P-orthophosphate for 2 h and 2 ng/ml PMA was added for the last 30 min of incubation. Cells were lysed, immunoprecipitated by anti-CD40, G28-5, (lanes A , B, E, and F) or anti-I-E, 13-4, as a control (lanes C and D) and analyzed by SDS-PAGE. The bands corresponding to CD40 are indicated by an arrow. Although the relative molecular mass of CD40 molecule was reported to be 50 kDa, in our hands 1251-or internally labeled CD40 was found to be much smaller (approximately 45 kDa). Recent biochemical analy3.3 CD40 expressed on transfectants is phosphorylated sis also showed that the relative molecular mass of CD40 was 43 or 44 kDa [34]. Cells used are: (Lanes A and C), unstimulated human CD40 is known to be a phosphorylated protein [33]. We tonsil E rosette-negative cells; (lanes B and D), PMA-stimulated examined whether CD40 expressed on the transfectants human tonsil E rosette-negative cells; (lane E), unstimulated was also phosphorylated. Cells were labeled with [32P]- CD40 expressing M12, K2C3; (lane F), PMA-stimulated CD40 orthophosphate in the presence or absence of PMA and expressing M12, K2C3. The relative molecular mass (kDa) of marker proteins is shown on the left of the gels.
Figure 3. The growth inhibitory effect of anti-CD40 on murine B and T transfectants expressing CD40. Experiments were done as described in Sect. 2.5. Anti-CD40, anti-CD23, anti-I-E and PMA were used at concentrations of 100 nglml, 250 ng/ml, 250 ng/ml and 2 ng/ml, respectively. The means k SD of [3H]dThd uptake obtained in triplicate experiments were shown.
immunoprecipitated by anti-CD40 mAb, G28-5. CD40 was constitutively phosphorylated in human tonsil B cells (Fig. 4, lane A). PMA enhanced CD40 phosphorylation in B cells (Fig. 4, lane B). In the CD40+ transfectant K2C3, CD40 was also phosphorylated without PMA and the phosphorylation was augmented by PMA (Fig. 4, lanes E and F). These results indicate that both CD40 molecules in normal B cells and transfectants are similarly phosphorylated.
3.4 The intracytoplasmic region essential for signal transduction through 0 4 0 CD40+ M12 cells were further studied to determine which part of the CD40 molecule was essential for signal transduction through CD40.We constructed two types of mutant 0 4 0 cDNA as shown schematically in Fig. 5. The first group included two mutants which contained deletions in the cytoplasmic region. The Del(201) mutant was made by introducing a TGA termination codon at position 202 by using a synthetic oligonucleotide linker. The other Del (232) mutant was made by a single-base substitution to change Gluu3 (GAG) into a termination codon (TAG).The Del (201) mutant and the Del (232) mutant encoded molecules, which had 6 and 37 amino acid residues in the cytoplasmic region, respectively. The second group included seven mutants encoding the molecules with single-amino acid substitutions in the cytoplasmic region. There are three serine, four threonine, and one cysteine residues in the intracellular region of CD40.The products of the six mutants, 203A(Ala203for Thr203),227A(Ala227for Thr227),234A(Alau4 for Thru4), 225A(Ala225for Ser225),
u -
l a
252 A
I
A
4
262
238G
I
-G ap8
Figure 5. Amino acid substitutions and deletions in the intracytoplasmic region of mutant CD40. (Upper part) Schematic structure of wild type CD40 cDNA. (Lower part) The positions of amino acid substitutions and deletions in the intracytoplasmic regions of mutant CD40 molecules. Corresponding amino acids in the wild type CD40 were also shown. The aminoterminal glutamic acid residue of mature CD40 determined by Braesch-Anderson et al. [21] is designated as + 1.
Signal transduction through CD40
Eur. J. Immunol. 1990. 20: 1747-1753
1751
4 Discussion The B cell differentiation antigen CD40 is thought to be the molecule involved in the regulation of human B cell growth, since certain agonistic antibodies, G28-5 [15, 161 and S2C6 [19], were shown to induce the proliferation of normal B cells in the presence of co-stimulants such as anti-IgM, PMA or anti-CD20. Here we show that the murine B lymphoma cell line M12 transfected with CD40 cDNA could respond to anti-CD40. In contrast to human B cells, the proliferation of CD40+ M12 was inhibited by antiCD40.The growth inhibitory effect was dose dependent for doses > 12.5 ng/ml. In the same assays, irrelevant antibodies employed as controls had no effect on either parental cell lines or transfectants. The concentrations of anti-CD40 required for the growth inhibition of such transfectants were very similar to those necessary for proliferation of normal B cells. Furthermore PMA augmented the antiCD40-induced growth signal and the phosphorylation of the CD40 molecule in transfectants as well as in normal B
The growth of cells 100 73
50 I
Figure 6. Expression of mutant CD40 in transfectants. M12 cells were transfected with various mutant CD40 cDNA. Repesentative clones for CD40+ transfectants were stained with FITC-avidin alone (---) or with biotinylated G28-5 (anti-CD40) plus FITCavidin (-) and analyzed by FCM. (A), Del(201); (B), Del(232);
(C), 203A; (D), 227A; (E), 234A; (F), 225A; (G), 249G; (H), 252A; (I), 238G.
3 f
1112
r i
K2C3 Dcl(201)
P u P I
2 4 9 G ( G l ~for~ ~ Ser249), ~ and 252A(Ala252for Ser252)were expected to contain single amino acid substitutions at the Ser or Thr residues in the cytoplasmic region. The mutant, 2 3 8 G ( G l for ~ ~ Cysus) ~~ was also made to change the only cysteine in the cytoplasmic region since this cysteine residue may be involved in the association with other molecules. All the mutant cDNAwere cloned into the same expression vector CDM8 as the wild type CD40 cDNA and were transfected into M12. Several clones expressing each mutant CD40 were isolated by several cycles of sorting.The CD40 expression levels of transfectants carrying mutant CD40 were similar to that of K2C3 (Fig. 6). Then PMAstimulated transfectants expressing mutant CD40 were compared for their responsiveness t o the antibody (Fig. 7). The products of 203A, 225A, 252A and 2386 could transduce the growth inhibitory signal of anti-CD40 as efficiently as an intact CD40 in PMA-stimulated transfectants. The inhibitory effect of anti-CD40 on transfectants expressing 227A and 2496 was rather weak. However, this inhibition was reproducible in four experiments with three different clones (data not shown). Among the mutant CD40 products with amino acid substitutions, only the products of the 234A failed to transduce the inhibitory signal by anti-CD40. The transfectants expressing mutant CD40 with deletions in the cytoplasmic region could not deliver a growth inhibitory signal, and all of them lacked Thr234.These results suggest that the cytoplasmic region around Thr234 might be essential in signal transduction through CD40.
Del(232)
203A 227A
234A 225A 2 4 ~ ; 252A
I
I
I
4
'L -
Figure 7. The growth inhibitory effect of anti-CD40 on transfectants expressing various mutant CD40. Parent cell line M12 and transfectants were cultured with 2 nglml PMA or with 2 ng/ml PMA plus 100 ng/ml anti-CD40 mAb. The growth of cells was expressed as the percentage of [3H]dThd uptake of unstimulated cells. Closed column and open column represents the growth of cells stimulated with PMA alone and with PMA plus anti-CD40 mAb, respectively. All experiments were carried out in triplicate.
1752
S. Inui,T. Kaisho, H. Kikutani et al.
cells. These results indicate that the CD40 expressed on murine cells is functional although it transduced an inhibitory signal. The growth of CD40-expressing EL4 cells was also inhibited by PMA and anti-CD40. This suggests that both murine B and Tcell lines may have a common signal transduction mechanism through CD40, which leads to growth inhibition in the transfectants. Why anti-CD40 induces the growth-inhibitory signal in transfectants is an interesting question. Similar findings were also observed in other systems. For example, the human CD25 (a chain of the IL2R) expressed on EL4 formed a high-affinity IL 2R with endogenous p chain, but transduced a growth inhibitory signal [35], while CD25 expressed on CTLL-2, which required IL 2 stimulation for its growth, transduced a growth-promoting one [36]. The proliferation of normal lymphocytes responding to growth stimulants is transient, suggesting that the negative feedback system would follow the growth-promoting events in such cells. In actively growing cells such as M12 or EL4, the growth-promoting effect of PMA or anti-CD40 is not observed and only the negative feedback system may be detectable in the assay system. The phosphorylation of EGF receptor by PMA was found to be important in the down-regulation of the receptor [37,38]. PMA caused the reduction of high-affinity EGF receptor by phosphorylating the EGF receptor and made cells less responsive to EGE In the present study, PMA enhanced phosphorylation of CD40 and accumulated phosphorylation of CD40 might then be involved in growth inhibition. However, we could not exclude the possibility that the different growth response between normal B cells and transfectants might result from the interaction of different molecules with CD40 after stimulation.
Eur. J. Immunol. 1990. 20: 1747-1753
phosphorylated and PMA enhanced this phosphorylation in the transfectants expressing 234A (data not shown) as well as in the transfectants expressing intact CD40, suggesting that serines or threonines other than Thr234 were phosphorylated in the transfectants. Presently we do not know whether Thr234itself is phosphorylated. Target amino acid residues for PKC are known to be often surrounded by basic amino acids such as lysine or arginine [39]. There are no basic amino acid residues around Thr234.Thus, further biochemical studies including phosphoamino acid analysis of CD40 will be necessary. Alternatively the region including Thr234might be involved in the association of CD40 with other molecules also involved in signal transduction. So far we have not identified any molecule, which B specifically associated with functional CD40, by immunoprecipitation using chemical cross-linking reagents or digitonin buffer (data not shown). Even if associated molecules do exist, the affinity of association may be too low or the molecular weight too small to identify them under the conditions used. Although CD40 is a surface molecule involved in B cell growth regulation, the natural ligand for CD40 has not been identified. The ligand may be a soluble factor, or a cell-to-cell interaction molecule. This transfectant system may provide a useful model to identify the natural ligand for CD40. We thank Ms. K. Kubota and M . Harayarna for their secretorial assistance.
Received March 26, 1990.
We made several mutant CD40 cDNA to study which part in the intracytoplasmic region of CD40 molecule is essential 5 References for its signal transduction. Transfectants expressing mutant 1 Kishimoto, T., Annu. Rev. Irnrnunol. 1985. 3: 133. CD40 were established and analyzed for the ability to 2 Kishimoto, T. and Hirano, T., Annu. Rev. Immunol. 1988. 6: transduce the growth inhibitory signal. Only three cell lines 485. expressing De1(201), Del(232) and 234A cDNA failed to 3 O’Garra, A . , Umland, S., DeFrance, T. and Christiansen, J., respond to anti-CD40 in the presence of PMA. This failure Irnrnunol. Today 1988. 9: 45. to transduce the signal was not due to low level expression 4 Paul,W., Cell 1989. 57: 521. of the CD40 epitope, although expression level of various 5 Clevers, H . , Alarcon, B. ,Wileman, T. and Terhorst, C., Annu. mutant CD40 was not exactly identical among each clone. Rev. Irnrnunol. 1988. 6: 629. First, the cells transfected with Del(201) and 234A (Fig. 6A 6 Meuer, S. C., Hussey, R. E . , Fabbi, M., Fox, D., Acuto, O., and E), which could not transmit any growth inhibitory Fitzgerald, K. A . , Hodgdon, J. C., Protentis, J. I?, Schlossman, S. F. and Reinherz, E. L., Cell 1984. 36: 897. signal, expressed the CD40 epitope as brightly as K2C3 7 Siliciano, R. F., Pratt, J. C., Schmidt, R. E., Ritz, J. and (Fig. 1A). Furthermore, the cells transfected with the 2386 Reinherz, E. L., Nature 1985. 317: 428. cDNA (Fig. 61), which was as active as K2C3 in terms of Fox, D. A . , Hussey, R. E., Fitzgerald, K. A . , Bensussan, A., 8 signal transduction, expressed lower level of CD40 epitope Daley, J. F., Schlossman, S. F. and Reinherz, E. L., J. Zmrnunol. than transfectants expressing intact CD40, De1(201), 1985. 134: 330. Del(232) and 234A (Figs. 1A and 6A, B, and E). All the 9 Plunkett, M. L., Sanders, M. E., Selvaraj,P., Dustin, M. L. and mutant CD40 molecules, which were inactive in terms of Springer,T. A., J. Exp. Med. 1987. 165: 664. signal transduction, lacked Thr234.Thisindicates that Thr234 10 Shaw, S . , Luce, G. E. G., Quinones, R., Gress, R. E ., Springer, itself or the region around Thr234is essential for the signal T. A. and Sanders, M. E., Nature 1986. 323: 262. 11 Pezzutto, A . , Dorken, B., Rabinovitch, P. S., Ledbetter, J. A., transduction through CD40.
Moldenhauer, G. and Clark, E. A . , J. Imrnunol. 1987. 138:
We compared the sequences around Thr234with cytoplasmic regions of other receptors, but found no significant homology with them. At the moment we do not know how this region contributes to the signal transduction mechanism through CD40. There are two possibilities. First, we must consider the possibility that Thr234might be the target for phosphorylation. The CD40 molecule was constitutively
2793. 12 Tedder,T.F., Streuli, M., Schlossman, S. F. and Saito, H., Proc. Natl. Acad. Sci. USA 1988. 85: 208. 13 Einfeld, D. A . , Brown, J. P.,Valentine,M. A . , Clark, E. A. and Ledbetter, J. A . , EMBO J. 1988. 7: 711. 14 Stamenkovic, I. and Seed, B., J. Exp. Med. 1988. 167: 1975. 15 Clark, E. A. and Ledbetter, J. A., Proc. Natl. Acad. Sci. USA 1986. 83: 4494.
Eur. J. Immunol. 1990. 20: 1747-1753 16 Ledbetter, J. A., Shu, G., Gallagher, M. and Clark, E. A., J. Immunol. 1987. 138: 788. 17 Stamenkovic, I., Clark, E. A. and Seed, B., EMBO J. 1989.8: 1403. 18 Hart, D. N. J. and McKenzie, J. L., J. Exp. Med. 1988. 168: 157. 19 Paulie, S., Ehlin-Henriksson, B., Mellstedt, H., Koho, H., Ben Aissa, H. and Perlmann, P., Cancer Immunol. Immunother. 1985. 20: 23. 20 Ledbetter, J. A., Clark, E. A., Noms, N. A., Shu, G. and Hellstroem, I., in McMichael, A. J. et al. (Eds.), Leucocyte Typing Ill, Oxford University Press, Oxford 1987, p. 432. 21 Braesch-Anderson, S., Paulie, S., Koho, H., Nika, H., Aspenstrom, P. and Perlmann, P., J. Immunol. 1989. 142: 562. 22 Ullrich, A., Coussens, L., Hayflick, J. S., DuII,T. J., Gray, A., Tam, A.W., Lee, J.,Yarden,Y., Libermann,T. A., Schlessinger, J., Downward, J., Mayes, E. L.V.,Whittle, N., Waterfield, M. D. and Seeburg, P. H., Nature 1984. 309: 418. 23 Coussens, L., Beveren, C.V, Smith, D., Chen, E., Mitchell, R. L., Isacke, C. M.,Verma, I. M. and Ullrich, A., Nature 1986. 320: 277. 24 Ullich, A., Bell, L. J. R., Chen, E.Y., Herrera, R., Petruzzelli, L. M., Dull, T. J., Gray, A., Coussens, L., Liao, Y.-C., Tsubokawa, M., Mason, A., Seeburg, P. H., Grunfeld, C., Rosen, 0. M. and Ramachandran, J., Nature 1985. 313: 756. 25 Johnson, D., Lanahan, A., Buck, C. R., Sehgal, A., Morgan, C., Mercer, E., Bothwell, M. and Chao, M., Cell 1986. 47: 545. 26 Hunter,T. and Cooper, J. A., Annu. Rev. Biochem. 1985. 54: 897. 27 Southern, I? J. and Berg, P., J. Mol. Appl. Genet. 1982. I : 327.
Signal transduction through CD40
1753
28 Kim, K. J., Kanellopoulos-Langevin, C., Merwin, R. M., Sachs, D. H. and Asofsky, R., J. Immunol. 1979. 122: 549. 29 Farrar, J. J., Fuller-Farrar, J., Simon, P. L., Hilfiker, M. L., Stadler, B. M. and Farrar, W. L., J. Immunol. 1980. 125: 2555. 30 Potter, H.,Weir, L. and Leder, F’., Proc. Natl. Acad. Sci. USA 1984. 81: 7161. 31 Sanger, F., Nicklen, S. and Coulson, A. R., Proc. Natl. Acad. Sci. USA 1977. 74: 5463. 32 Suemura, M., Kikutani, H., Barsumian, E. L., Hattori, Y., Kishimoto, S., Sato, R., Maeda, A., Nakamura, H., Owaki, H., Hardy, R. R. and Kishimoto,T., J. Immunol. 1986. 137: 1214. 33 Paulie, S., RosCn, A., Ehlin-Henriksson, B., Braesch-Andersen, S., Jakobson, E., Koho, H. and Perlmann, P., J. Immunol. 1989. 142: 590. 34 Ling, N. R., MacLennan, I. C. M. and Mason, D. Y. in McMichael, A. J. et al. (Eds.), Leucocyte Typing I l l , Oxford University Press, Oxford 1987, p. 302. 35 Hatakeyama, M., Minamoto, S., Uchiyama,T., Hardy, R. R., Yamada, G. and lhniguchi, T., Nature 1985. 318: 467. 36 Kondo, S., Shimizu, A., Maeda, M.,Tagaya,Y.,Yodoi, J. and Honjo, T., Nature 1986. 320: 75. 37 Livneh, E., Reiss, N., Berent, E., Ullrich, A. and Schlessinger, J., EMBO J. 1987. 6: 2669. 38 Lin, C. R., Chen,W. S., Lazar, C. S., Carpenter, C. D., Gill, G. N., Evans, R. M. and Rosenfeld, M. G., Cell 1986. 44: 839. 39 Woodgett, J. R., Gould, K. L. and Hunter,T., Eur. J. Biochem. 1986. 161: 177.
Eur. J. Immunol. 1990. 20: 1747-1753
Seiji Inuiv, Tsuneyasu Kaisho, Hitoshi Kikutani, Ivan StamenkovicAo, Brian SeedAn, Edward A. Clarko and Tadamitsu Kishimoto Division of Immunology, Institute for Molecular and Cellular Biology, Osaka University, Osaka, Department of Geneticso, Harvanl Medical School and Department of Molecular BiologyA, Massachusetts General Hospital, Boston and Department of Microbiologyo, University of Washington, Seattle
1747
Identification of the intracytoplasmic region essential for signal transduction through a B cell activation molecule, CDN* CD40 is a 45-kDa glycoprotein expressed on human B lineage cells. Anti-CD40 induces the proliferation of B cells and the extracellular region of CD40 is related to those of a certain kind of growth factor receptors. Therefore, it has been proposed that CD40 might be a receptor for a molecule involved in the growth regulation of B cells.The cDNA coding for CD40 was transfected into the murine B lymphoma cell line M12 and the murine thymoma cell line EL4.The growth of both M12 and EL4 transfectants expressing human CD40 was inhibited by anti-CD40. Phorbol 12-myristate 13-acetate (PMA) augmented the growth inhibitory effects of anti-CD40 on transfectants. The CD40 molecule was constitutively phosphorylated not only in human tonsil B cells but also in transfectants expressing CD40. PMA augmented the phosphorylation of CD40 in these cells. These results indicate that in spite of the growth inhibitory effect of anti-CD40, the augmentative effect of PMA is conserved in CD40+ transfectants and suggest that the transfectant might be useful for the study of signal transduction mechanism through CD40. To investigate which part of the CD40 molecule is important for signal transduction, transfectants expressing mutant CD40 cDNA were established and their growth response to anti-CD40 was evaluated.The mutant molecule, which had an Ala for Thr substitution at position 234, and the deletion mutants lacking Thr234were inactive in growth signal transduction, indicating that Thr234itself or the region around Thr234is essential for signal transduction through CD40.
1 Introduction A number of lymphokines or cytokines involved in the regulation of B cell growth or differentiation have been identified, molecularly cloned and shown to have a pleiotropic effect on various cells including T cells, MWmonocytes as well as on B cells [l-41. The receptors for these molecules are expressed not only on B cells but also on other lineage cells. On the other hand, there certainly exist some lineage-restricted surface molecules which are involved in cell activation, growth or differentiation. Tcells express the CD3 complex and the CD2 molecule.The CD3 complex is associated with the TcR and transduces the signals necessary for Tcell activation [5]. Certain anti-CD2 antibodies were shown to induce the proliferation of Tcells [6-81. Moreover, the interaction of CD2 with another cell surface molecule, the lymphocyte function-associated antigen 3 (LFA-3), was found to play an important role in antigen-independent cell-to-cell communication [9, 101. Certain B lineage-restricted membrane molecules such as CD19, CD20 or CD40 have also been proposed to be
involved in B cell activation or growth [ll-171, although their physiological roles remain to be resolved. CD40 is a 45-kDa glycoprotein specifically expressed on the surface of B lineage cells and dendritic cells of hematopoietic cell origin [15, 181. CD40 was also shown to be expressed on tumor cells, mostly of epithelial origin [16, 19-21]. Expression of CD40 was enhanced when B cells were activated with PMA or anti-IgM [16]. The proliferation of normal B cells could be induced by anti-CD40 when they were preactivated with anti-IgM, PMA, or anti-CD20 [15]. These results indicate that CD40 is one of growth factor receptors involved both in tumor growth and in normal B cell proliferation.
The CD40 cDNA was cloned by Stamenkovic et al. [17] using a high-efficiency expression system in mammalian cells. These sequence analysis revealed that CD40 has 20 cysteine residues in the extramembrane region while only one in the cytoplasmic region. The receptors of a certain kind of growth factors such as epidermal growth factor (EGF), CSF-1, and insulin also have cysteine-rich extramembrane regions. In fact, the positions of the cysteine residues are conserved in the extramembrane regions of [I 84471 CD40 and the nerve growth factor (NGF) receptor [22-251. * This study was supported in part by Grant-in-Aid for Specially However, no tyrosine kinase domain, which is considered Promoted Research from the Ministry of Education, Science crucial for signal transduction in a certain family of and Culture and by National Institutes of Health Grant receptors [26], was observed in the cytoplasmic region of GM37905. CD40 or NGF receptor [17, 251. Thus CD40 appears to Present address: Basel Institute for Immunology, CH-4005 belong to one of the families of growth factor receptors that Basel, Switzerland use signal transduction mechanisms differing from those Correspondence: Tadamitsu Kishimoto, Institute for Molecular with a tyrosine kinase domain.
and Cellular Biology, Osaka University, 1-3-Yamada-oka, Osaka 565, Japan Abbreviations: growth factor
EGF Epidermal growth factor NGF Nerve
0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
To elucidate the function of the signal transduction mechanisms through CD40, we transfected CD40 cDNA into murine cell lines. Anti-CD40 and PMA inhibited the proliferation of transfectants expressing CD40 and PMA 0014-2980/90/0808-1747$3.50+ .25/0
1748
Eur. J. Immunol. 1990.20: 1747-1753
S. Inui,T. Kaisho, H. Kikutani et al.
enhanced the phosphorylation of CD40 in transfectants as well as in normal human B cells. These results suggest that the CD40 molecule on transfectants can transduce the growth signal and that the transfectants might be useful for the analysis of signal transduction mechanisms. Using this transfectant system and mutant CD40 cDNA, we further showed that the cytoplasmic region around Thr234was essential for signal transduction through CD40.
2 Materials and methods 2.1 DNA The CD40 cDNA was cloned by Stamenkovic et al. [17] with an expression vector (CDM8) carrying the transcription unit composed of the human cytomegalovirus enhancer and the HIV long terminal repeat. The plasmid pSV2neo carries the neomycin resistance gene and imparted resistance to G418 [27]. Twenty microgram of wild type or mutant CD40 cDNA and 2 pg of pSV2neo were linearized by Sac I1 and Barn HI, respectively, and employed for transfection of 1 x 107cells.
5’-GAGATGCGACCCTCCC-3r(Ser2494 Gly249), 5’-CCTGCACTGCGATGCGACTC-3‘(Ser252 4 AlaZ2), 5 ’-CCGGTTGGCCTCCATGTAAA-~’(CYS~ + Gly’*), 5’- GTA A AGTCTACTGCACTGG A-3' (G ~ +uSTOP), ~ ~ and yielded the mutant cDNA 203A, 227A, 234A, 225A, 249G, 252A, 2386 and Del (232), respectively. The plasmids containing mutant CD40 cDNA were isolated and sequenced around the modified region by the dideoxy sequencing method [31]. The double-stranded DNA were prepared from E. coli infected with each M13mp19 mutant clone and digested with Hind I11 and Sma I. The resulting DNA fragments were subcloned into CDM8 which had been previously prepared. Another deletion mutant Del (201) (Pro202 + STOP) was made as follows. The CD40 cDNA was digested with Xho I and Bal I, and ligated to the linkers, 5’-CCAGAAGTGACTGCA-3’ and 5’-GTCACTTCTTGG-3‘, to change Pro202into a termination codon, and provide a Bal I restriction recognition site. The resulting fragment was subcloned into the Pst I and Xho I sites of CDM8. 2.4 Cell staining and sorting
2.2 Transfection of DNA DNA was transfected into the murine B lymphoma cell line M12 [28] and the thymoma cell line EL4 [29] by the electroporation method [30]. Ten million exponentially growing cells were harvested, washed twice with PBS, and then resuspended at a concentration of 1 x lo7 cells/ml in ice-cold PBS (120.7 mM KC1, 30.8 mM NaC1, 8.1 mM Na2HP04, 1.46 mM KH2P04) including 5 mM MgC12. Plasmid DNA (see above) were added to the cell suspension and kept on ice for at least 15 min. An electric pulse of 1.5 kV was applied by a Promega X cell-2000 power supply (Seikagaku Kogyo, Co., Ltd.,Tokyo, Japan).The cells were kept on ice for 7 min, transfered to the RPMI 1640 medium supplemented with 10% FCS and kept at room temperature for 10 min. After two days of culture in normal medium, M12 and EL4 transfectants were selected in 1.0 and 0.5 mg/ml G418 (Gibco Laboratories, Grand Island, NY), respectively.
One million cells were incubated with biotinylated G28-5 (anti-CD40; [15]) for 20 rnin at 4 ° C washed three times with staining buffer (PBS, 2% FCS, 0.02% NaN3), and counterstained with FITC-conjugated avidin. Propidium iodide was included for the last 5 min to gate out the data provided by dead cells. Stained cells were analyzed and sorted on a fluorescence activated cell sorter (FACS 440, Becton Dickinson, Mountain View, CA). 2.5 Proliferation assay Cells were cultured at 1 x 104/wellin 96-well microtiter plates containing 200 p1 of RPMI 1640 medium supplemented with 2% FCS, antibiotics, L-glutamine and 2-ME with or without various stimulations. After 3 days, cells were pulsed with 0.5 pCi = 18.5 kBq of [3H]thymidine/well for the last 3 h. Cells were then harvested with a cell harvester onto glass-fiber filters, and the radioactivity was measured.
2.3 Construction of mutant CD40 cDNA 2.6 The phosphorylation of CD40 To construct mutated CD40 molecules, the plasmid CDM8 carrying the CD40 cDNA was digested with restriction enzyme Not I, blunted with Klenow fragment of E. coli DNA polymerase I and further digested with Hind 111.The CD40 cDNA blunt-Hind 111 fragment was inserted into M13mp19 previously digested with Hind 111and Hinc I1 and employed to produce a single-stranded DNA template for oligonucleotide-directed mutagenesis. Oligonucleotidedirected mutagenesis was performed using an in vitro mutagenesis kit (Amersham Int., Amersham, GB). All oligonucleotides used were synthesized on 381A DNA synthesizer (Applied Biosystems, Fostercity, CA). Singlebase substitutions were introduced using appropriate mismatched 20-mer oligonucleotides as follows: 5’-CCTTAlTGGCTGGCITCTTG-3’(Thr203 + Ala203)),
Ten million actively growing cells were washed three times by phosphate-free RPMI 1640 medium and incubated at 1 x 106/ml for 2 h in phosphate-free RPMI 1640 medium supplemented with 10% FCS. For E rosette-negative cells, 1 x los cells were used and incubated at 1 x 107/ml.Cells were labeled with 1mCi [”2P]orthophosphate in 2 ml for 2 h at 37°C. For the last 30 min, PMA was added at 2 ng/ml. Cells were lysed with 0.4 ml of a solution containing 50 mM Hepes, pH 7.4, 1% Triton X-100,lO mM sodium pyrophosphate, 100 mM NaF, 4 mM EDTA, 2 mM sodium orthovanadate, 1 mM PMSF and 100 U/ml aprotinin. Cell lysates were clarified by centrifuging at 14000 x g for 60 rnin at 4°C. SN were precleared with an irrevalent antibody, 3-5 [30], and protein G-Sepharose (Pharmacia, Piscataway, NJ). Immunoprecipitation was carried out by G28-5 and 5‘-GAGCAGCAGCG?TGGAGCCA-3’(Thr227 + 5r-CATGTAAAGCCTCCTGCACT-3f(Thr234 + Ala234), protein G-Sepharose. Immunoprecipitates were washed 5’-AGCAGTGTTGGCGCCAGGAA-3’(Ser225 + Ala2=), four times with a washing solution containig 50 mM Hepes,
~
Signal transduction through CD40
Eur. J. Immunol. 1990. 20: 1747-1753
1749
pH 7.4, 1% Triton X-100, 0.1% SDS, 150 mM NaCl, 100 mM NaF and 2 mM sodium orthovanadate, and eluted by boiling in electrophoresis sample buffer (2% SDS, 10% glycerol, 5% 2-ME and 62.5 mM Tris-HCI, pH 6.8). Samples were analyzed by SDS-PAGE, and subjected to autoradiography.
3 Results 3.1 Establishment of murine cell lines expressing CD40
The murine B lymphoma cell line M12 was transfected with the CD40 cDNA and pSV2neo. After two weeks of G418 selection, drug-resistant clones were obtained and screened by indirect immunofluorescence as described in Sect. 2.4. After several cycles of sorting, the expression level of CD40 became stable without further sorting. FCM analysis pattern of a representative CD40+ clone, K2C3, is shown in Fig. 1A. This clone was employed in the following functional analysis. Similarly, the murine thymoma cell line EL4 was transfected with CD40 cDNA and CD40+ clone 2B1 was established (Fig. 1B). 3.2 Functional analysis of CD40- expressing transfectants Anti-CD40 was shown to induce the proliferation of normal B cells in the presence of co-stimulants such as anti-IgM, PMA, or anti-CD20 [15]. K2C3 was stimulated by various concentrations of anti-CD40 in the absence or presence of PMA to analyze whether the CD40 expressed on the murine cells could transduce a growth signal (Fig. 2). In contrast to normal human B cells, anti-CD40 inhibited the proliferation of K2C3. The growth inhibitory effect of anti-CD40 was dose dependent and observed at doses as low as 12.5 ng/ml. Anti-CD40 inhibited the growth of K2C3
(BI
0.1
1
10
100 0.1 Fluorescein (log)
1
10
100
Figure 1. FCM analysis of CD40-expressing transfectants. Parent cell lines (M12 and EL4) and transfectants (CD40-expressing M12, K2C3 and CD40-expressing ELI, 2B1) were stained with biotinylated G28-5 (anti-CD40) and FITC-conjugated avidin. (A) M12(-) and K2C3(-); (B) ELI(-) and 2B1(-). Ten thousand cells were analyzed per sample on the FACS 440. Histograms show relative cell number (y-axis) vs. log fluorescence (x-axis). Dotted line(---) is a fluorescence profile of the transfectants stained with FITCconjugated avidin alone.
b
io 160 Antibody concentration
ng/ml
Figure 2. The growth inhibitory effect of various concentrations of anti-CD40 on the murine B cell line M12 expressing CD40. Ten thousand cells were cultured for 3 days with various concentrations of anti-CD40 in the presence or absence of PMA. Proliferation was measured as the incorporation of [3H]dThd during the last 3 h. MlZ(0); MI2 with 2 ng/ml PMA(0); K2C3(0); K2C3 with 2 nglml PMA(W). Results were expressed as the percentage of [3H]dThd uptake of cells with no stimulation. All experiments were carried out in triplicate cultures.
up to 90% in the presence of PMA. The proliferation of other CD40+ transfectant clones such as K2C4, which expresses lower levels of CD40 than K2C3, was also inhibited similarly by anti-CD40 and PMA (data not shown). However, antLCD40 or PMA had very little effect on the parental M12 cells even when added together. The antibody concentrations required for growth inhibition of K2C3 were similar to those required for augmenting the proliferation of normal human B cells [15, 161. Although PMA alone also did inhibit the growth of transfectants, such an inhibitory effect is variable among transfectants (data not shown). In spite of variable effects by PMA alone, PMA consistently augmented the growth inhibition by anti-CD40. To test the specificity of anti-CD40-induced growth inhibition, we examined the effect of control antibodies, either antLCD23 (3-5; [32]) which was of the same isotype (IgGI) as anti-CD40, or anti-I-E (13/4) which reacted with both transfectants and parental cells. Control antibodies at 250 ng/ml showed no further inhibitory effect on the proliferation of PMA-stimulated transfectants (Fig. 3A). These results demonstrate that the growth inhibition of K2C3 is CD40 specific. To study whether the growth inhibitory signal is B cell specific, the effect of anti-CD40 on CD40+ murine T cell transfectant was analyzed. Likewise the proliferation of 2B1 derived from EL4 was inhibited by antLCD40 and PMA (Fig. 3B). Neither antLCD23 nor anti-Thy-1.2 inhibited the proliferation of EL4 and 2B1 (data not shown).These results indicate that the growth of transfectants are negatively regulated by the signal through CD40 in both murine B and Tcell lines.
1750
S. Inui,T. Kaisho, H. Kikutani et al.
Eur. J. Immunol. 1990. 20: 1747-1753
Figure 4 . CD40 molecules expressed on transfectants as well as normal human B cells are phosphorylated. Cells were labeled with 32P-orthophosphate for 2 h and 2 ng/ml PMA was added for the last 30 min of incubation. Cells were lysed, immunoprecipitated by anti-CD40, G28-5, (lanes A , B, E, and F) or anti-I-E, 13-4, as a control (lanes C and D) and analyzed by SDS-PAGE. The bands corresponding to CD40 are indicated by an arrow. Although the relative molecular mass of CD40 molecule was reported to be 50 kDa, in our hands 1251-or internally labeled CD40 was found to be much smaller (approximately 45 kDa). Recent biochemical analy3.3 CD40 expressed on transfectants is phosphorylated sis also showed that the relative molecular mass of CD40 was 43 or 44 kDa [34]. Cells used are: (Lanes A and C), unstimulated human CD40 is known to be a phosphorylated protein [33]. We tonsil E rosette-negative cells; (lanes B and D), PMA-stimulated examined whether CD40 expressed on the transfectants human tonsil E rosette-negative cells; (lane E), unstimulated was also phosphorylated. Cells were labeled with [32P]- CD40 expressing M12, K2C3; (lane F), PMA-stimulated CD40 orthophosphate in the presence or absence of PMA and expressing M12, K2C3. The relative molecular mass (kDa) of marker proteins is shown on the left of the gels.
Figure 3. The growth inhibitory effect of anti-CD40 on murine B and T transfectants expressing CD40. Experiments were done as described in Sect. 2.5. Anti-CD40, anti-CD23, anti-I-E and PMA were used at concentrations of 100 nglml, 250 ng/ml, 250 ng/ml and 2 ng/ml, respectively. The means k SD of [3H]dThd uptake obtained in triplicate experiments were shown.
immunoprecipitated by anti-CD40 mAb, G28-5. CD40 was constitutively phosphorylated in human tonsil B cells (Fig. 4, lane A). PMA enhanced CD40 phosphorylation in B cells (Fig. 4, lane B). In the CD40+ transfectant K2C3, CD40 was also phosphorylated without PMA and the phosphorylation was augmented by PMA (Fig. 4, lanes E and F). These results indicate that both CD40 molecules in normal B cells and transfectants are similarly phosphorylated.
3.4 The intracytoplasmic region essential for signal transduction through 0 4 0 CD40+ M12 cells were further studied to determine which part of the CD40 molecule was essential for signal transduction through CD40.We constructed two types of mutant 0 4 0 cDNA as shown schematically in Fig. 5. The first group included two mutants which contained deletions in the cytoplasmic region. The Del(201) mutant was made by introducing a TGA termination codon at position 202 by using a synthetic oligonucleotide linker. The other Del (232) mutant was made by a single-base substitution to change Gluu3 (GAG) into a termination codon (TAG).The Del (201) mutant and the Del (232) mutant encoded molecules, which had 6 and 37 amino acid residues in the cytoplasmic region, respectively. The second group included seven mutants encoding the molecules with single-amino acid substitutions in the cytoplasmic region. There are three serine, four threonine, and one cysteine residues in the intracellular region of CD40.The products of the six mutants, 203A(Ala203for Thr203),227A(Ala227for Thr227),234A(Alau4 for Thru4), 225A(Ala225for Ser225),
u -
l a
252 A
I
A
4
262
238G
I
-G ap8
Figure 5. Amino acid substitutions and deletions in the intracytoplasmic region of mutant CD40. (Upper part) Schematic structure of wild type CD40 cDNA. (Lower part) The positions of amino acid substitutions and deletions in the intracytoplasmic regions of mutant CD40 molecules. Corresponding amino acids in the wild type CD40 were also shown. The aminoterminal glutamic acid residue of mature CD40 determined by Braesch-Anderson et al. [21] is designated as + 1.
Signal transduction through CD40
Eur. J. Immunol. 1990. 20: 1747-1753
1751
4 Discussion The B cell differentiation antigen CD40 is thought to be the molecule involved in the regulation of human B cell growth, since certain agonistic antibodies, G28-5 [15, 161 and S2C6 [19], were shown to induce the proliferation of normal B cells in the presence of co-stimulants such as anti-IgM, PMA or anti-CD20. Here we show that the murine B lymphoma cell line M12 transfected with CD40 cDNA could respond to anti-CD40. In contrast to human B cells, the proliferation of CD40+ M12 was inhibited by antiCD40.The growth inhibitory effect was dose dependent for doses > 12.5 ng/ml. In the same assays, irrelevant antibodies employed as controls had no effect on either parental cell lines or transfectants. The concentrations of anti-CD40 required for the growth inhibition of such transfectants were very similar to those necessary for proliferation of normal B cells. Furthermore PMA augmented the antiCD40-induced growth signal and the phosphorylation of the CD40 molecule in transfectants as well as in normal B
The growth of cells 100 73
50 I
Figure 6. Expression of mutant CD40 in transfectants. M12 cells were transfected with various mutant CD40 cDNA. Repesentative clones for CD40+ transfectants were stained with FITC-avidin alone (---) or with biotinylated G28-5 (anti-CD40) plus FITCavidin (-) and analyzed by FCM. (A), Del(201); (B), Del(232);
(C), 203A; (D), 227A; (E), 234A; (F), 225A; (G), 249G; (H), 252A; (I), 238G.
3 f
1112
r i
K2C3 Dcl(201)
P u P I
2 4 9 G ( G l ~for~ ~ Ser249), ~ and 252A(Ala252for Ser252)were expected to contain single amino acid substitutions at the Ser or Thr residues in the cytoplasmic region. The mutant, 2 3 8 G ( G l for ~ ~ Cysus) ~~ was also made to change the only cysteine in the cytoplasmic region since this cysteine residue may be involved in the association with other molecules. All the mutant cDNAwere cloned into the same expression vector CDM8 as the wild type CD40 cDNA and were transfected into M12. Several clones expressing each mutant CD40 were isolated by several cycles of sorting.The CD40 expression levels of transfectants carrying mutant CD40 were similar to that of K2C3 (Fig. 6). Then PMAstimulated transfectants expressing mutant CD40 were compared for their responsiveness t o the antibody (Fig. 7). The products of 203A, 225A, 252A and 2386 could transduce the growth inhibitory signal of anti-CD40 as efficiently as an intact CD40 in PMA-stimulated transfectants. The inhibitory effect of anti-CD40 on transfectants expressing 227A and 2496 was rather weak. However, this inhibition was reproducible in four experiments with three different clones (data not shown). Among the mutant CD40 products with amino acid substitutions, only the products of the 234A failed to transduce the inhibitory signal by anti-CD40. The transfectants expressing mutant CD40 with deletions in the cytoplasmic region could not deliver a growth inhibitory signal, and all of them lacked Thr234.These results suggest that the cytoplasmic region around Thr234 might be essential in signal transduction through CD40.
Del(232)
203A 227A
234A 225A 2 4 ~ ; 252A
I
I
I
4
'L -
Figure 7. The growth inhibitory effect of anti-CD40 on transfectants expressing various mutant CD40. Parent cell line M12 and transfectants were cultured with 2 nglml PMA or with 2 ng/ml PMA plus 100 ng/ml anti-CD40 mAb. The growth of cells was expressed as the percentage of [3H]dThd uptake of unstimulated cells. Closed column and open column represents the growth of cells stimulated with PMA alone and with PMA plus anti-CD40 mAb, respectively. All experiments were carried out in triplicate.
1752
S. Inui,T. Kaisho, H. Kikutani et al.
cells. These results indicate that the CD40 expressed on murine cells is functional although it transduced an inhibitory signal. The growth of CD40-expressing EL4 cells was also inhibited by PMA and anti-CD40. This suggests that both murine B and Tcell lines may have a common signal transduction mechanism through CD40, which leads to growth inhibition in the transfectants. Why anti-CD40 induces the growth-inhibitory signal in transfectants is an interesting question. Similar findings were also observed in other systems. For example, the human CD25 (a chain of the IL2R) expressed on EL4 formed a high-affinity IL 2R with endogenous p chain, but transduced a growth inhibitory signal [35], while CD25 expressed on CTLL-2, which required IL 2 stimulation for its growth, transduced a growth-promoting one [36]. The proliferation of normal lymphocytes responding to growth stimulants is transient, suggesting that the negative feedback system would follow the growth-promoting events in such cells. In actively growing cells such as M12 or EL4, the growth-promoting effect of PMA or anti-CD40 is not observed and only the negative feedback system may be detectable in the assay system. The phosphorylation of EGF receptor by PMA was found to be important in the down-regulation of the receptor [37,38]. PMA caused the reduction of high-affinity EGF receptor by phosphorylating the EGF receptor and made cells less responsive to EGE In the present study, PMA enhanced phosphorylation of CD40 and accumulated phosphorylation of CD40 might then be involved in growth inhibition. However, we could not exclude the possibility that the different growth response between normal B cells and transfectants might result from the interaction of different molecules with CD40 after stimulation.
Eur. J. Immunol. 1990. 20: 1747-1753
phosphorylated and PMA enhanced this phosphorylation in the transfectants expressing 234A (data not shown) as well as in the transfectants expressing intact CD40, suggesting that serines or threonines other than Thr234 were phosphorylated in the transfectants. Presently we do not know whether Thr234itself is phosphorylated. Target amino acid residues for PKC are known to be often surrounded by basic amino acids such as lysine or arginine [39]. There are no basic amino acid residues around Thr234.Thus, further biochemical studies including phosphoamino acid analysis of CD40 will be necessary. Alternatively the region including Thr234might be involved in the association of CD40 with other molecules also involved in signal transduction. So far we have not identified any molecule, which B specifically associated with functional CD40, by immunoprecipitation using chemical cross-linking reagents or digitonin buffer (data not shown). Even if associated molecules do exist, the affinity of association may be too low or the molecular weight too small to identify them under the conditions used. Although CD40 is a surface molecule involved in B cell growth regulation, the natural ligand for CD40 has not been identified. The ligand may be a soluble factor, or a cell-to-cell interaction molecule. This transfectant system may provide a useful model to identify the natural ligand for CD40. We thank Ms. K. Kubota and M . Harayarna for their secretorial assistance.
Received March 26, 1990.
We made several mutant CD40 cDNA to study which part in the intracytoplasmic region of CD40 molecule is essential 5 References for its signal transduction. Transfectants expressing mutant 1 Kishimoto, T., Annu. Rev. Irnrnunol. 1985. 3: 133. CD40 were established and analyzed for the ability to 2 Kishimoto, T. and Hirano, T., Annu. Rev. Immunol. 1988. 6: transduce the growth inhibitory signal. Only three cell lines 485. expressing De1(201), Del(232) and 234A cDNA failed to 3 O’Garra, A . , Umland, S., DeFrance, T. and Christiansen, J., respond to anti-CD40 in the presence of PMA. This failure Irnrnunol. Today 1988. 9: 45. to transduce the signal was not due to low level expression 4 Paul,W., Cell 1989. 57: 521. of the CD40 epitope, although expression level of various 5 Clevers, H . , Alarcon, B. ,Wileman, T. and Terhorst, C., Annu. mutant CD40 was not exactly identical among each clone. Rev. Irnrnunol. 1988. 6: 629. First, the cells transfected with Del(201) and 234A (Fig. 6A 6 Meuer, S. C., Hussey, R. E . , Fabbi, M., Fox, D., Acuto, O., and E), which could not transmit any growth inhibitory Fitzgerald, K. A . , Hodgdon, J. C., Protentis, J. I?, Schlossman, S. F. and Reinherz, E. L., Cell 1984. 36: 897. signal, expressed the CD40 epitope as brightly as K2C3 7 Siliciano, R. F., Pratt, J. C., Schmidt, R. E., Ritz, J. and (Fig. 1A). Furthermore, the cells transfected with the 2386 Reinherz, E. L., Nature 1985. 317: 428. cDNA (Fig. 61), which was as active as K2C3 in terms of Fox, D. A . , Hussey, R. E., Fitzgerald, K. A . , Bensussan, A., 8 signal transduction, expressed lower level of CD40 epitope Daley, J. F., Schlossman, S. F. and Reinherz, E. L., J. Zmrnunol. than transfectants expressing intact CD40, De1(201), 1985. 134: 330. Del(232) and 234A (Figs. 1A and 6A, B, and E). All the 9 Plunkett, M. L., Sanders, M. E., Selvaraj,P., Dustin, M. L. and mutant CD40 molecules, which were inactive in terms of Springer,T. A., J. Exp. Med. 1987. 165: 664. signal transduction, lacked Thr234.Thisindicates that Thr234 10 Shaw, S . , Luce, G. E. G., Quinones, R., Gress, R. E ., Springer, itself or the region around Thr234is essential for the signal T. A. and Sanders, M. E., Nature 1986. 323: 262. 11 Pezzutto, A . , Dorken, B., Rabinovitch, P. S., Ledbetter, J. A., transduction through CD40.
Moldenhauer, G. and Clark, E. A . , J. Imrnunol. 1987. 138:
We compared the sequences around Thr234with cytoplasmic regions of other receptors, but found no significant homology with them. At the moment we do not know how this region contributes to the signal transduction mechanism through CD40. There are two possibilities. First, we must consider the possibility that Thr234might be the target for phosphorylation. The CD40 molecule was constitutively
2793. 12 Tedder,T.F., Streuli, M., Schlossman, S. F. and Saito, H., Proc. Natl. Acad. Sci. USA 1988. 85: 208. 13 Einfeld, D. A . , Brown, J. P.,Valentine,M. A . , Clark, E. A. and Ledbetter, J. A . , EMBO J. 1988. 7: 711. 14 Stamenkovic, I. and Seed, B., J. Exp. Med. 1988. 167: 1975. 15 Clark, E. A. and Ledbetter, J. A., Proc. Natl. Acad. Sci. USA 1986. 83: 4494.
Eur. J. Immunol. 1990. 20: 1747-1753 16 Ledbetter, J. A., Shu, G., Gallagher, M. and Clark, E. A., J. Immunol. 1987. 138: 788. 17 Stamenkovic, I., Clark, E. A. and Seed, B., EMBO J. 1989.8: 1403. 18 Hart, D. N. J. and McKenzie, J. L., J. Exp. Med. 1988. 168: 157. 19 Paulie, S., Ehlin-Henriksson, B., Mellstedt, H., Koho, H., Ben Aissa, H. and Perlmann, P., Cancer Immunol. Immunother. 1985. 20: 23. 20 Ledbetter, J. A., Clark, E. A., Noms, N. A., Shu, G. and Hellstroem, I., in McMichael, A. J. et al. (Eds.), Leucocyte Typing Ill, Oxford University Press, Oxford 1987, p. 432. 21 Braesch-Anderson, S., Paulie, S., Koho, H., Nika, H., Aspenstrom, P. and Perlmann, P., J. Immunol. 1989. 142: 562. 22 Ullrich, A., Coussens, L., Hayflick, J. S., DuII,T. J., Gray, A., Tam, A.W., Lee, J.,Yarden,Y., Libermann,T. A., Schlessinger, J., Downward, J., Mayes, E. L.V.,Whittle, N., Waterfield, M. D. and Seeburg, P. H., Nature 1984. 309: 418. 23 Coussens, L., Beveren, C.V, Smith, D., Chen, E., Mitchell, R. L., Isacke, C. M.,Verma, I. M. and Ullrich, A., Nature 1986. 320: 277. 24 Ullich, A., Bell, L. J. R., Chen, E.Y., Herrera, R., Petruzzelli, L. M., Dull, T. J., Gray, A., Coussens, L., Liao, Y.-C., Tsubokawa, M., Mason, A., Seeburg, P. H., Grunfeld, C., Rosen, 0. M. and Ramachandran, J., Nature 1985. 313: 756. 25 Johnson, D., Lanahan, A., Buck, C. R., Sehgal, A., Morgan, C., Mercer, E., Bothwell, M. and Chao, M., Cell 1986. 47: 545. 26 Hunter,T. and Cooper, J. A., Annu. Rev. Biochem. 1985. 54: 897. 27 Southern, I? J. and Berg, P., J. Mol. Appl. Genet. 1982. I : 327.
Signal transduction through CD40
1753
28 Kim, K. J., Kanellopoulos-Langevin, C., Merwin, R. M., Sachs, D. H. and Asofsky, R., J. Immunol. 1979. 122: 549. 29 Farrar, J. J., Fuller-Farrar, J., Simon, P. L., Hilfiker, M. L., Stadler, B. M. and Farrar, W. L., J. Immunol. 1980. 125: 2555. 30 Potter, H.,Weir, L. and Leder, F’., Proc. Natl. Acad. Sci. USA 1984. 81: 7161. 31 Sanger, F., Nicklen, S. and Coulson, A. R., Proc. Natl. Acad. Sci. USA 1977. 74: 5463. 32 Suemura, M., Kikutani, H., Barsumian, E. L., Hattori, Y., Kishimoto, S., Sato, R., Maeda, A., Nakamura, H., Owaki, H., Hardy, R. R. and Kishimoto,T., J. Immunol. 1986. 137: 1214. 33 Paulie, S., RosCn, A., Ehlin-Henriksson, B., Braesch-Andersen, S., Jakobson, E., Koho, H. and Perlmann, P., J. Immunol. 1989. 142: 590. 34 Ling, N. R., MacLennan, I. C. M. and Mason, D. Y. in McMichael, A. J. et al. (Eds.), Leucocyte Typing I l l , Oxford University Press, Oxford 1987, p. 302. 35 Hatakeyama, M., Minamoto, S., Uchiyama,T., Hardy, R. R., Yamada, G. and lhniguchi, T., Nature 1985. 318: 467. 36 Kondo, S., Shimizu, A., Maeda, M.,Tagaya,Y.,Yodoi, J. and Honjo, T., Nature 1986. 320: 75. 37 Livneh, E., Reiss, N., Berent, E., Ullrich, A. and Schlessinger, J., EMBO J. 1987. 6: 2669. 38 Lin, C. R., Chen,W. S., Lazar, C. S., Carpenter, C. D., Gill, G. N., Evans, R. M. and Rosenfeld, M. G., Cell 1986. 44: 839. 39 Woodgett, J. R., Gould, K. L. and Hunter,T., Eur. J. Biochem. 1986. 161: 177.
