Low expression of the Tcell receptorKD3 complex
Eur. J. Immunol. 1991. 21: 1641-1647
1641
Franqoise Le Deist, Gabriela Thoenes, Jose Corado, Barbara Lisowska-Grospierre and Main Fischer
Immunodeficiency with low expression of the T cell receptodCD3 complex. Effect on T lymphocyte activation*
INSERM Ul32,HBpital Necker-Enfants Malades, Paris
We report the consequences of low expression of the Tcell receptor (TcR)/CD3 complex by T lymphocytes from a 4-year-old boy with a mild immunodeficiency. TcR/CD3 expression was found to be deficient on both resting and activated T cells, using both anti-CD3 and anti-TcRdP monoclonal antibodies. As shown by immunofluorescence and immunoprecipitation studies, residual expression (corresponding to about 10% of normal) was detectable on resting and activated TcR a@+ Tcells. Other Tcell membrane receptors were normally expressed. The functional consequences of this TcR/CD3 expression deficiency included an absence of Tcell proliferation, interleukin 2 receptor expression and calcium flux following anti-CD3 and anti-CD2 antibody-triggered T cell activation. Antigen (tetanus toxoid, Candida and allogeneic cell)-induced proliferation was detectable. In contrast, cytotoxic Tcell activity towards allogeneic cells was deficient. These findings shed light on the function of the TcR/CD3 complex and indicate that the expression of a limited number of TcR/CD3 receptors may be sufficient to trigger antigen-specific T cell activation (and, possibly, differentiation) and that anti-CD3 antibody-induced T cell activation differs somewhat from antigen/ major histocompatibility complex molecule-induced activation .These results also confirm that the CD2 pathway of T cell activation is CD3 dependent.
1 Introduction An increasing number of primary Tcell deficiencies related to defective T cell differentiation or abnormal membrane receptor expression have been described [1-41. Defective expression of the TcR/CD3 complex has been reported in two siblings, one with severe immunodeficiency and the other with normal immune function [5,6]. The TcR/CD3 complex normally consists of either the a and p chains or the y and 6 chains of the TcR noncovalently bound to CD3y, 6, E, 5 and possibly q proteins [7-9].TheTcR itself is the recognition element, while the CD3 subunits might transduce the activation signal triggered by antigen/MHC recognition 17, 81. It was found that immunodeficiency with low TcR/CD3 expression was associated with poor biosynthesis of the CD3c chain which may lead to abnormal expression of an incomplete complex [6]. Such studies have provided invaluable information on the biology of TcR/CD3 complex formation, since the apparent deficiency of the CD3 5 chain is associated with the expression of an immature non-glycosylated TcR/CD3 complex [6]. Such T cells cannot be activated by anti-CD3 antibody or antigens, while anti-CD2 antibody-induced activation occurs. We report a further case of a child with mild cellular and humoral immunodeficiency and a partial defect in T cell
expression of the TcR/CD3 complex. The unknown primary defect may be distinct from that underlying the low TcR/CD3 expression found in the Spanish family, since biosynthesis of CD3 I; is normal [lo]. We now describe the immunodeficiency and its consequences on Tcell activation and function.
2 Materials and methods 2.1 Case report
I? T., the 2-year-old son of unrelated parents presented recurrent Hemophilus influenzae pneumonia and otitis media over a 6-month period. He was started on prophylactic antibiotic therapy and i.v. Ig and, 2years later, remains infection free. Blood cell counts have remained within normal values (lymphocyte counts 2600 to 4500/pl) . Skin DTH tests to PHAwere negative. Serum Ig levels were elevated at time of chronic infection (IgG 29 mg/ml, IgA 13 mg/ml and IgM 6.3 mg/ml), later returning to normal values. No monoclonal Ig components were detected. Antibody production was partially defective, since no antibodies to poliovirus or isohemagglutinin were detected (normal values 3 10 x lo-' UI/ml, titer 3 8 x lo-*, respectively). In contrast, serum antibody levels to tetanus and diphtheria toxoids were within the normal range (1 UI/ml and 1UI/ml, respectively: normal values >0.1 and > 0.1 UI/ml). Erythrocyte adenosine deaminase and purine nucleotide phosphorylase activities were normal.
[I 93411
*
This study was supported by INSERM.
2.2 Cell isolation
Correspondence: FranGoise Le Deist, INSERM U 132, HGpital des PBMC were isolated from freshly drawn heparinized blood Enfants Malades, 149. rue de SBvres, F-75743 Paris Cedex 14, by means of Ficoll-Hypaque (Pharmacia Fine Chemicals, Uppsala, Sweden) density gradient centrifugation. France 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
oO14-2980/91/0707-161$3.50+ .25/0
1642
F. Le Deist, G. Thoenes, J. Corado et al.
Eur. J. Immunol. 1991. 21: 1641-1647
2.3 Surface marker analysis
(final dilution 11200). UV-excited indo-1 fluorescence was split into high- and low-wavelength emission with a 450 nm Fluorescence staining was performed on either PBMC or T long-pass dichroic mirror. The reflected short-wavelength cell blasts. The following mAb were used: anti-CD3: Leu-4 emissions were passed throug a 405/20-nm band-pass filter; (IgG2,; Becton Dickinson, Mountain View, CA), IOT3 long-wavelength emissions were passed through a (IgG2,; Immunotech, Marseille, France), OKT3 (IgG1; 485/42-nm band-pass filter. The 405 nm : 485 nm fluoresOrtho Pharmaceutical, Raritan, NJ), anti-TcR a/@: cence ratio was displayed as a function of time. BMA031 (IgG1; Behring-Werke, Marburg/Lahn, FRG), TcR 1(IgG1; Becton Dickinson), anti-TcR 6 :TcR 61 (IgG1; T cell Sciences, Cambridge, MA), anti-CD2: Leu-5b 2.6 Cytotoxicity assay (IgG2,; Becton Dickinson), anti-CD4: Leu-3a (IgG1; Becton Dickinson), anti-CD8: Leu-2a (IgG1; Becton Dickin- CTL activity was tested in a 4-h standard Cr-release assay, son), anti-CD5: Leu-1 (IgG2,; Becton Dickinson), anti- as previously described [17]. The effector cells were alloCD18: IOT18 (IgG1; Immunotech), anti-CD16: Leu-llb geneic cell-stimulated PBMC and cytotoxic activity was (IgGI; Becton Dickinson), anti-CD45RO: UCHLl (IgG2,; tested against PHA (Difco; v : v = MOO) 3-day-stimulated kindly provided by I? Beverley), anti-CD45RA: 2H4 (IgGl; allogeneic cells. In the NK assay, the effector cells were Coulter Clone, Margency, France), anti-CD29: 4B4 (IgGl; PBMC and the target cells were K-562 cells.The results are Coulter Clone), anti-HLA class I: IOT2 (IgG2,; Immuno- expressed as cytotoxic indices (CI) calculated according to tech), anti-HLA class I1 DR: (IgG2,; Becton Dickinson), following formula: anti-CD25: rIL 2 (IgG1; Becton Dickinson), anti-memSample release - spontaneous release brane IgM (Nordic, Tilburg, The Netherlands). CI = x loo. Maximal release - spontaneous release
Direct fluorescence staining was performed using PE- or FITC-conjugated mAb. An FITC-conjugated goat antimouse Ig (Nordic) was used for indirect immunofluores- 2.7 TcR Cg gene rearrangement cence. The analysis was performed using a FACStar plus (Becton Dickinson). The logarithm scale was verified using Total genomic DNA was extracted from PBMC. Ten a CEM clone (which expresses 5000 to 10000 CD5 mole- micrograms of DNA was digested to completion by selected cules) and fluorescein-calibrated beads (Becton Dickin- restriction endonucleases (Bam HI, Eco RI or Hind 111). The restriction fragments were size-fractionated by electroson). phoresis on 0.9% agarose gels and blotted onto nylon filters (Hybond H+, Amersham, Les Ulis, France) in 0 . 4 ~ NaOH. The filters were screened using a 770-bp cDNA Cp 2.4 Cell cultures region probe of the gene [MI. The probe was radioactively Proliferation assays were performed as described elsewhere labeled by random priming (Boehringer-Mannheim, MeyIll]. Briefly, PBMC were stimulated either with mitogens lan, France) and the filter hybridized in 50% formamide at in 4-day cultures or with antigen and allogeneic cells in 42 "C. This Cp probe recognizes Cgl and Cp2 regions of the 6-day cultures. OKT3 antibody (50 ng/ml, 500 ng/ml), two TcR gene and their rearrangement can be observed with different described anti-CD2 antibody pairs, D66 plus Eco RI and Hind I11 digests, respectively. T1l.l antibody ([12]; final dilution 1/200) kindly provided by A. Bernard, and CD2.1 plus CD2.9 ([13]; final dilution 1/1500)kindly provided by D. Olive, PHA (Difco, Detroit, 2.8 Analysis of iodinated surface-expressedTcWCD3 complexes MI; final dilution 1/700), and the association of PMA (Sigma, St. Louis, MO) and ionomycin (Calbiochem, San Diego, CA) were used as mitogens. Anti-CD28 (mAb 248 PBMC of the patient and the control were surface iodinated kindly provided by Daniel Olive; [14]; final dilution 1/500) by lZ5I using Iodogen (tetrachloro-diphenylglycoluril; was also used in conjunction with OKT3 [15]. SpecificTcell Pierce, Cheshire, GB)-coated vials with 37 MBq for proliferation was driven by tetanus toxoid (Pasteur Diag- 10 X lo6 cells as described by Laemmli [19]. The cells were nostic, Marne la Coquette, France) or candidin (Pasteur then lysed in digitonin/Triton lysis buffer. The TcR/CD3 Diagnostic) at the appropriate concentrations, and by complexes were immunoprecipitated by the anti-TcR PF1 (Tcell Sciences, Boston, MA) plus the anti-CD3 antibody allogeneic cells. Leu-4 (Becton Dickinson; 5 pl). SDS-PAGE was performed first under nonreducing (NR) , then under reducing (R) conditions. Autoradiograms were exposed for 21 and 2.5 Assay of intracellular calcium concentration 3 days for patient and control samples, respectively. PBMC were loaded with indo-l-acetoxymethylester (indo-1, AM; Molecular Probes, Eugene, OR) as described by Breitmeyer et al. [16]. Briefly, 10 x lo6 PBMC were 3 Results incubated for 30 min at 37°C in indo-1 (4 pM). The cells were washed, then further incubated for 1h. The indo-1 3.1 Membrane expression of TcWCD3 fluorescence ratio of individual cells, and indicator of the intracellular free calcium concentration, was measured The patient's lymphocytes were tested for CD3 membrane using a FACStar Plus before and after the addition of OKT3 expression using three CD3 &-specificmAb: Leu-4, IOT3 antibody (125 ng/ml) plus rabbit anti-mouse Ig (RaMIg; (data not shown) and OKT3. Alow level of CD3 expression Nordic; final dilution 1/100)or the anti-CD2 antibodies pair was consistently found except on 2% of T cells which
Eur. J. Immunol. 1991. 21: 1641-1647
Low expression of the Tcell receptorKD3 complex
I
Figure 1. Expression of TcR/CD3 complex on the patient's lymphocytes. Reactivity of various mAb was analyzed by direct immunofluorescence (Leu-4, TcR 61) or indirect immunofluorescence (OKT3, BMA031, TcR 1). Fluorescence intensity was expressed on a logarithmic scale. Expression on the patient's lymphocytes (-) was compared to expression on control lymphocytes (---). For TcRG1, the acquisition was only performed on cells with a fluorescence intensity > 200.
B rl
u
z
3E
FLUORESCENCE INTENSITY (log scale)
expressed normal levels (Fig. 1). Based on the histogram which has a 1024 channel logarithm scale for four decades, we can estimate the patient's lymphocyte CD3 expression as tenfold lower than normal. The accuracy of this estimation was confirmed using a CEM clone which expresses 5000 to 10000 CD5 molecules and fluorescein-calibrated beads (data not shown). TcR a@ expression, studied with either TcR 1o r BMA031 antibody was also low (Fig. 1). No Tcells expressing the TcR 6 chains were detected using the TcR61 antibody (Fig. 1). This could reflect either an absence of 6+ T cells or low-level expression of these
1643
molecules on a small number of cells. The latter possibility is especially difficult to rule out, since theTcR61 antibody gave lower fluorescence intensity on normal cells than that obtained with anti-CD3 antibody or BMA031 (Fig. 1). Biochemical analysis of iodinated TcR/CD3 complexes showed a faint spot that may correspond to the y, 6 and E subunits and another on the diagonal at approximately 45 kDa (Fig. 2). In contrast, y, 6 and E subunits as well as T c R d p subunits were precipitated from control T cell membranes by the PF1 and Leu-4 antibodies. TcR/CD3 complex expression was also low on candidin-induced blasts (Fig. 1). In contrast, with resting T lymphocytes no Tcell blasts with normal CD3 expression were detected. Doublelabeling experiments confirmed that the cells with low PATIENT
CONTROL
FLUORESCENCE INTENSITY (log scale)
Figure 2. Analysis of cell surface-iodinated TcWCD3 complexes of the patient and a control. PBMC of the patient and a control were surface iodinated by lZI and lysed in digitonflriton lysis buffer. The TcIUCD3 complexes were immunoprecipitated by the anti-TcR antibody pF1 plus the anti-CD3 antibody Leu-4. The immunoprecipitates were resolved first under nonreducing (NR) and then reducing (R) conditions. Autoradiograms were exposed for 21 and 3 days for patient and control, respectively.
Figure 3. Expression of CD2, CD3, CD4 on the patient's lymphocytes. CD2, CD3, CD4 expression was tested respectively with FITC- or PE-conjugated Leu-Sb, Leu-4 and Leu-3a by direct immunofluorescence. Fluorescence intensity was expressed on a log scale.The double CD3 (Y axis)/CD2 (X axis) fluorescence was studied in (A) and (B), and the double CD3 (Y axis)/CD4 (X axis) fluorescence was studied in (C) and (D); patient's lymphocytes (A and C) and control lymphocytes (B and D).
1644
Eur. J. Immunol. 1991. 21: 1641-1647
F. Le Deist, G. Thoenes, J. Corado et al.
Table 1. Cell surface antigen expression
Antigen tested
CD3 CD4 CD8 CD2 CD18
CD45RO(UCHLl) CD45RA (2H4) CD29 (4B4) CD16 Membrane IgM HLA class I HLA-DR on B lymphocytes
Percentage of positive cells
discrete rearranged bands were present (data not shown). Mean fluorescence intensity (logarithmic scale) Patient Control
Patient
Control
40
82
338
22 33 78 70 39 20 50 35 6
50 32 83
500
524 470 620 624
581 573 569 416 643
92
25 22 16 28 10 10 83
412 519 675 722
335 410 466 617 664
78
88
574
571
355
608
TcR/CD3 expression were normal T cells (CD4+ for instance) and not other cell types such as NK (Fig. 3). The expression of other surface molecules (Table 1) including CD2, CD4, CD8, CD45 isotypes, CD29 and CD18 o n T lymphocytes was normal. However, it is noticeable that the percentage of CD4+Tcells was diminished. The proportion of B cells appeared to be normal, with no anomaly in the expression of sIg. An excess of NK (CD16) cells was detected (Table 1). Lymphocytes from the patient’s healthy sister and parents showed normal expression of the TcR/CD3 complex (data not shown). Cg TcR element rearrangement was studied using a Cp probe. The rearrangement pattern was normal, since no
3.2 Functional studies The patient’s T lymphocytes failed to proliferate in the presence of PHA, OKT3 and the anti-CD2 antibody pairs (Table 2), even at an OKT3 antibody concentration as high as 500 ng/ml (Table 2). The absence of T cell proliferation was associated with a low increase in the cytosolic free calcium concentration after activation either by OKT3 antibody or by an antLCD2 antibody pair (Fig. 4) and with a very low expression of membrane activation antigens (Table 3). Interestingly, co-stimulation with an antLCD28 antibody restored anti-CD3 induced T cell activation (Table 2). However, the patient’s cells proliferated in the presence of PMA plus ionomycin. In contrast, both antigens tested (tetanus toxoid, candidin) induced T cell proliferation (Table 2). Surprisingly, dose-response curves were not different from control curves (Fig. 5). Proliferation in response to allogeneic cells was also normal (Table 2), but cytotoxicTcel1 activity was absent (Table 4). NK cytotoxic activity against the cell line K-562 was high, in agreement with the increased number of CD16 cells (Table 4).
4 Discussion In this report, we describe the occurrence of low TcR/CD3 complex expression on peripheral Tcells from a child with a recurrent infection. The deficiency was repeatedly found over a 2-year period, using various anti-CD3 &-specific antibodies and two TcR a@-specific antibodies both by immunofluorescence and immunoprecipitation studies. TcR/CD3 expression was approximately 10% of normal.
Figure 4. Calcium flux followingTcel1activation. Indo-1-loaded patient (A and C) or control lymphocytes (B and D) were activated either with anti-CD3 mAb (125 ng/ml) plus RaMIg (final dilution 1/100) (A and B) or with anti-CD2 pair [D66 +T11.1; final dilution 1/200; (C and D)]. Samples were monitored discontinuously for 3 min and 30 s in (A) and (B) and for 7 min and 30 s in (C) and (D). Results are expressed as 405/485 nm fluorescence ratio in arbitrary units as a function of time.
Low expression of the T cell receptorlCD3 complex
Eur. J. Immunol. 1991. 21: 1641-1647
1645
Table 2. Lymphocyte proliferationa) Patient
Stimulus
+
m
800f
None (day 4) PHA OKT3 (Snglml) OKT3 (500ng/ml) OKT3 (50ng/d) i-mAb 248 Anti-CD2 ( 0 2 . 1 0 2 . 9 ) PMA (lW9 M) Anti-CD2 (D661+T11.1) PMA ( M)+ ionomycin (lo-* M) None (day 6) Tktanus toxoid
Normal values (cpm x 10-3)
700+ 300 126000f37000 36oOof23oOo 57000f27oOo 225000f35000 110000f 5 5 m 157000f 33000 168ooO+ 62000 800f loo 31000 f 19ooO 34000f21600 35000f 17400
loo0
2300+ 21m+ 16000f
400 600
500 300 9000f 2000
+
1000f
78800+15000 1m+ 300 22500+ 6900 23mf 6400 11500f 2500
candidin Allogeneic cells
a) Results are expressed as mean incorporation of [3H]dThd (cpm 1SD). b) Anti-CD28 antibody (mAb 248) alone did not induce T cell proliferation.
*
Table 3. Expression of activation antigens
P e t c a w of positive
Mean fluorescence
ah
intensity (Iog d e ) Patient Control
.
+
PMA
Patient
Control
76 34 2 8
82 36
cD25 HLA-DR
ionomycin M) OKT3 (SO nglml)
cD25
HLA-DR
487
-
53 40
-
NK activity Patient
50/1
74 67 60 28
2511 131 611
Cl’L activity Normal values
Normal values
Patient
53 k 21
2 2 1
n+9 17 f 7 11 26
ND
ND
41 k 19 97k16 17-r-13
Eur. J. Immunol. 1991. 21: 1641-1647
1641
Franqoise Le Deist, Gabriela Thoenes, Jose Corado, Barbara Lisowska-Grospierre and Main Fischer
Immunodeficiency with low expression of the T cell receptodCD3 complex. Effect on T lymphocyte activation*
INSERM Ul32,HBpital Necker-Enfants Malades, Paris
We report the consequences of low expression of the Tcell receptor (TcR)/CD3 complex by T lymphocytes from a 4-year-old boy with a mild immunodeficiency. TcR/CD3 expression was found to be deficient on both resting and activated T cells, using both anti-CD3 and anti-TcRdP monoclonal antibodies. As shown by immunofluorescence and immunoprecipitation studies, residual expression (corresponding to about 10% of normal) was detectable on resting and activated TcR a@+ Tcells. Other Tcell membrane receptors were normally expressed. The functional consequences of this TcR/CD3 expression deficiency included an absence of Tcell proliferation, interleukin 2 receptor expression and calcium flux following anti-CD3 and anti-CD2 antibody-triggered T cell activation. Antigen (tetanus toxoid, Candida and allogeneic cell)-induced proliferation was detectable. In contrast, cytotoxic Tcell activity towards allogeneic cells was deficient. These findings shed light on the function of the TcR/CD3 complex and indicate that the expression of a limited number of TcR/CD3 receptors may be sufficient to trigger antigen-specific T cell activation (and, possibly, differentiation) and that anti-CD3 antibody-induced T cell activation differs somewhat from antigen/ major histocompatibility complex molecule-induced activation .These results also confirm that the CD2 pathway of T cell activation is CD3 dependent.
1 Introduction An increasing number of primary Tcell deficiencies related to defective T cell differentiation or abnormal membrane receptor expression have been described [1-41. Defective expression of the TcR/CD3 complex has been reported in two siblings, one with severe immunodeficiency and the other with normal immune function [5,6]. The TcR/CD3 complex normally consists of either the a and p chains or the y and 6 chains of the TcR noncovalently bound to CD3y, 6, E, 5 and possibly q proteins [7-9].TheTcR itself is the recognition element, while the CD3 subunits might transduce the activation signal triggered by antigen/MHC recognition 17, 81. It was found that immunodeficiency with low TcR/CD3 expression was associated with poor biosynthesis of the CD3c chain which may lead to abnormal expression of an incomplete complex [6]. Such studies have provided invaluable information on the biology of TcR/CD3 complex formation, since the apparent deficiency of the CD3 5 chain is associated with the expression of an immature non-glycosylated TcR/CD3 complex [6]. Such T cells cannot be activated by anti-CD3 antibody or antigens, while anti-CD2 antibody-induced activation occurs. We report a further case of a child with mild cellular and humoral immunodeficiency and a partial defect in T cell
expression of the TcR/CD3 complex. The unknown primary defect may be distinct from that underlying the low TcR/CD3 expression found in the Spanish family, since biosynthesis of CD3 I; is normal [lo]. We now describe the immunodeficiency and its consequences on Tcell activation and function.
2 Materials and methods 2.1 Case report
I? T., the 2-year-old son of unrelated parents presented recurrent Hemophilus influenzae pneumonia and otitis media over a 6-month period. He was started on prophylactic antibiotic therapy and i.v. Ig and, 2years later, remains infection free. Blood cell counts have remained within normal values (lymphocyte counts 2600 to 4500/pl) . Skin DTH tests to PHAwere negative. Serum Ig levels were elevated at time of chronic infection (IgG 29 mg/ml, IgA 13 mg/ml and IgM 6.3 mg/ml), later returning to normal values. No monoclonal Ig components were detected. Antibody production was partially defective, since no antibodies to poliovirus or isohemagglutinin were detected (normal values 3 10 x lo-' UI/ml, titer 3 8 x lo-*, respectively). In contrast, serum antibody levels to tetanus and diphtheria toxoids were within the normal range (1 UI/ml and 1UI/ml, respectively: normal values >0.1 and > 0.1 UI/ml). Erythrocyte adenosine deaminase and purine nucleotide phosphorylase activities were normal.
[I 93411
*
This study was supported by INSERM.
2.2 Cell isolation
Correspondence: FranGoise Le Deist, INSERM U 132, HGpital des PBMC were isolated from freshly drawn heparinized blood Enfants Malades, 149. rue de SBvres, F-75743 Paris Cedex 14, by means of Ficoll-Hypaque (Pharmacia Fine Chemicals, Uppsala, Sweden) density gradient centrifugation. France 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
oO14-2980/91/0707-161$3.50+ .25/0
1642
F. Le Deist, G. Thoenes, J. Corado et al.
Eur. J. Immunol. 1991. 21: 1641-1647
2.3 Surface marker analysis
(final dilution 11200). UV-excited indo-1 fluorescence was split into high- and low-wavelength emission with a 450 nm Fluorescence staining was performed on either PBMC or T long-pass dichroic mirror. The reflected short-wavelength cell blasts. The following mAb were used: anti-CD3: Leu-4 emissions were passed throug a 405/20-nm band-pass filter; (IgG2,; Becton Dickinson, Mountain View, CA), IOT3 long-wavelength emissions were passed through a (IgG2,; Immunotech, Marseille, France), OKT3 (IgG1; 485/42-nm band-pass filter. The 405 nm : 485 nm fluoresOrtho Pharmaceutical, Raritan, NJ), anti-TcR a/@: cence ratio was displayed as a function of time. BMA031 (IgG1; Behring-Werke, Marburg/Lahn, FRG), TcR 1(IgG1; Becton Dickinson), anti-TcR 6 :TcR 61 (IgG1; T cell Sciences, Cambridge, MA), anti-CD2: Leu-5b 2.6 Cytotoxicity assay (IgG2,; Becton Dickinson), anti-CD4: Leu-3a (IgG1; Becton Dickinson), anti-CD8: Leu-2a (IgG1; Becton Dickin- CTL activity was tested in a 4-h standard Cr-release assay, son), anti-CD5: Leu-1 (IgG2,; Becton Dickinson), anti- as previously described [17]. The effector cells were alloCD18: IOT18 (IgG1; Immunotech), anti-CD16: Leu-llb geneic cell-stimulated PBMC and cytotoxic activity was (IgGI; Becton Dickinson), anti-CD45RO: UCHLl (IgG2,; tested against PHA (Difco; v : v = MOO) 3-day-stimulated kindly provided by I? Beverley), anti-CD45RA: 2H4 (IgGl; allogeneic cells. In the NK assay, the effector cells were Coulter Clone, Margency, France), anti-CD29: 4B4 (IgGl; PBMC and the target cells were K-562 cells.The results are Coulter Clone), anti-HLA class I: IOT2 (IgG2,; Immuno- expressed as cytotoxic indices (CI) calculated according to tech), anti-HLA class I1 DR: (IgG2,; Becton Dickinson), following formula: anti-CD25: rIL 2 (IgG1; Becton Dickinson), anti-memSample release - spontaneous release brane IgM (Nordic, Tilburg, The Netherlands). CI = x loo. Maximal release - spontaneous release
Direct fluorescence staining was performed using PE- or FITC-conjugated mAb. An FITC-conjugated goat antimouse Ig (Nordic) was used for indirect immunofluores- 2.7 TcR Cg gene rearrangement cence. The analysis was performed using a FACStar plus (Becton Dickinson). The logarithm scale was verified using Total genomic DNA was extracted from PBMC. Ten a CEM clone (which expresses 5000 to 10000 CD5 mole- micrograms of DNA was digested to completion by selected cules) and fluorescein-calibrated beads (Becton Dickin- restriction endonucleases (Bam HI, Eco RI or Hind 111). The restriction fragments were size-fractionated by electroson). phoresis on 0.9% agarose gels and blotted onto nylon filters (Hybond H+, Amersham, Les Ulis, France) in 0 . 4 ~ NaOH. The filters were screened using a 770-bp cDNA Cp 2.4 Cell cultures region probe of the gene [MI. The probe was radioactively Proliferation assays were performed as described elsewhere labeled by random priming (Boehringer-Mannheim, MeyIll]. Briefly, PBMC were stimulated either with mitogens lan, France) and the filter hybridized in 50% formamide at in 4-day cultures or with antigen and allogeneic cells in 42 "C. This Cp probe recognizes Cgl and Cp2 regions of the 6-day cultures. OKT3 antibody (50 ng/ml, 500 ng/ml), two TcR gene and their rearrangement can be observed with different described anti-CD2 antibody pairs, D66 plus Eco RI and Hind I11 digests, respectively. T1l.l antibody ([12]; final dilution 1/200) kindly provided by A. Bernard, and CD2.1 plus CD2.9 ([13]; final dilution 1/1500)kindly provided by D. Olive, PHA (Difco, Detroit, 2.8 Analysis of iodinated surface-expressedTcWCD3 complexes MI; final dilution 1/700), and the association of PMA (Sigma, St. Louis, MO) and ionomycin (Calbiochem, San Diego, CA) were used as mitogens. Anti-CD28 (mAb 248 PBMC of the patient and the control were surface iodinated kindly provided by Daniel Olive; [14]; final dilution 1/500) by lZ5I using Iodogen (tetrachloro-diphenylglycoluril; was also used in conjunction with OKT3 [15]. SpecificTcell Pierce, Cheshire, GB)-coated vials with 37 MBq for proliferation was driven by tetanus toxoid (Pasteur Diag- 10 X lo6 cells as described by Laemmli [19]. The cells were nostic, Marne la Coquette, France) or candidin (Pasteur then lysed in digitonin/Triton lysis buffer. The TcR/CD3 Diagnostic) at the appropriate concentrations, and by complexes were immunoprecipitated by the anti-TcR PF1 (Tcell Sciences, Boston, MA) plus the anti-CD3 antibody allogeneic cells. Leu-4 (Becton Dickinson; 5 pl). SDS-PAGE was performed first under nonreducing (NR) , then under reducing (R) conditions. Autoradiograms were exposed for 21 and 2.5 Assay of intracellular calcium concentration 3 days for patient and control samples, respectively. PBMC were loaded with indo-l-acetoxymethylester (indo-1, AM; Molecular Probes, Eugene, OR) as described by Breitmeyer et al. [16]. Briefly, 10 x lo6 PBMC were 3 Results incubated for 30 min at 37°C in indo-1 (4 pM). The cells were washed, then further incubated for 1h. The indo-1 3.1 Membrane expression of TcWCD3 fluorescence ratio of individual cells, and indicator of the intracellular free calcium concentration, was measured The patient's lymphocytes were tested for CD3 membrane using a FACStar Plus before and after the addition of OKT3 expression using three CD3 &-specificmAb: Leu-4, IOT3 antibody (125 ng/ml) plus rabbit anti-mouse Ig (RaMIg; (data not shown) and OKT3. Alow level of CD3 expression Nordic; final dilution 1/100)or the anti-CD2 antibodies pair was consistently found except on 2% of T cells which
Eur. J. Immunol. 1991. 21: 1641-1647
Low expression of the Tcell receptorKD3 complex
I
Figure 1. Expression of TcR/CD3 complex on the patient's lymphocytes. Reactivity of various mAb was analyzed by direct immunofluorescence (Leu-4, TcR 61) or indirect immunofluorescence (OKT3, BMA031, TcR 1). Fluorescence intensity was expressed on a logarithmic scale. Expression on the patient's lymphocytes (-) was compared to expression on control lymphocytes (---). For TcRG1, the acquisition was only performed on cells with a fluorescence intensity > 200.
B rl
u
z
3E
FLUORESCENCE INTENSITY (log scale)
expressed normal levels (Fig. 1). Based on the histogram which has a 1024 channel logarithm scale for four decades, we can estimate the patient's lymphocyte CD3 expression as tenfold lower than normal. The accuracy of this estimation was confirmed using a CEM clone which expresses 5000 to 10000 CD5 molecules and fluorescein-calibrated beads (data not shown). TcR a@ expression, studied with either TcR 1o r BMA031 antibody was also low (Fig. 1). No Tcells expressing the TcR 6 chains were detected using the TcR61 antibody (Fig. 1). This could reflect either an absence of 6+ T cells or low-level expression of these
1643
molecules on a small number of cells. The latter possibility is especially difficult to rule out, since theTcR61 antibody gave lower fluorescence intensity on normal cells than that obtained with anti-CD3 antibody or BMA031 (Fig. 1). Biochemical analysis of iodinated TcR/CD3 complexes showed a faint spot that may correspond to the y, 6 and E subunits and another on the diagonal at approximately 45 kDa (Fig. 2). In contrast, y, 6 and E subunits as well as T c R d p subunits were precipitated from control T cell membranes by the PF1 and Leu-4 antibodies. TcR/CD3 complex expression was also low on candidin-induced blasts (Fig. 1). In contrast, with resting T lymphocytes no Tcell blasts with normal CD3 expression were detected. Doublelabeling experiments confirmed that the cells with low PATIENT
CONTROL
FLUORESCENCE INTENSITY (log scale)
Figure 2. Analysis of cell surface-iodinated TcWCD3 complexes of the patient and a control. PBMC of the patient and a control were surface iodinated by lZI and lysed in digitonflriton lysis buffer. The TcIUCD3 complexes were immunoprecipitated by the anti-TcR antibody pF1 plus the anti-CD3 antibody Leu-4. The immunoprecipitates were resolved first under nonreducing (NR) and then reducing (R) conditions. Autoradiograms were exposed for 21 and 3 days for patient and control, respectively.
Figure 3. Expression of CD2, CD3, CD4 on the patient's lymphocytes. CD2, CD3, CD4 expression was tested respectively with FITC- or PE-conjugated Leu-Sb, Leu-4 and Leu-3a by direct immunofluorescence. Fluorescence intensity was expressed on a log scale.The double CD3 (Y axis)/CD2 (X axis) fluorescence was studied in (A) and (B), and the double CD3 (Y axis)/CD4 (X axis) fluorescence was studied in (C) and (D); patient's lymphocytes (A and C) and control lymphocytes (B and D).
1644
Eur. J. Immunol. 1991. 21: 1641-1647
F. Le Deist, G. Thoenes, J. Corado et al.
Table 1. Cell surface antigen expression
Antigen tested
CD3 CD4 CD8 CD2 CD18
CD45RO(UCHLl) CD45RA (2H4) CD29 (4B4) CD16 Membrane IgM HLA class I HLA-DR on B lymphocytes
Percentage of positive cells
discrete rearranged bands were present (data not shown). Mean fluorescence intensity (logarithmic scale) Patient Control
Patient
Control
40
82
338
22 33 78 70 39 20 50 35 6
50 32 83
500
524 470 620 624
581 573 569 416 643
92
25 22 16 28 10 10 83
412 519 675 722
335 410 466 617 664
78
88
574
571
355
608
TcR/CD3 expression were normal T cells (CD4+ for instance) and not other cell types such as NK (Fig. 3). The expression of other surface molecules (Table 1) including CD2, CD4, CD8, CD45 isotypes, CD29 and CD18 o n T lymphocytes was normal. However, it is noticeable that the percentage of CD4+Tcells was diminished. The proportion of B cells appeared to be normal, with no anomaly in the expression of sIg. An excess of NK (CD16) cells was detected (Table 1). Lymphocytes from the patient’s healthy sister and parents showed normal expression of the TcR/CD3 complex (data not shown). Cg TcR element rearrangement was studied using a Cp probe. The rearrangement pattern was normal, since no
3.2 Functional studies The patient’s T lymphocytes failed to proliferate in the presence of PHA, OKT3 and the anti-CD2 antibody pairs (Table 2), even at an OKT3 antibody concentration as high as 500 ng/ml (Table 2). The absence of T cell proliferation was associated with a low increase in the cytosolic free calcium concentration after activation either by OKT3 antibody or by an antLCD2 antibody pair (Fig. 4) and with a very low expression of membrane activation antigens (Table 3). Interestingly, co-stimulation with an antLCD28 antibody restored anti-CD3 induced T cell activation (Table 2). However, the patient’s cells proliferated in the presence of PMA plus ionomycin. In contrast, both antigens tested (tetanus toxoid, candidin) induced T cell proliferation (Table 2). Surprisingly, dose-response curves were not different from control curves (Fig. 5). Proliferation in response to allogeneic cells was also normal (Table 2), but cytotoxicTcel1 activity was absent (Table 4). NK cytotoxic activity against the cell line K-562 was high, in agreement with the increased number of CD16 cells (Table 4).
4 Discussion In this report, we describe the occurrence of low TcR/CD3 complex expression on peripheral Tcells from a child with a recurrent infection. The deficiency was repeatedly found over a 2-year period, using various anti-CD3 &-specific antibodies and two TcR a@-specific antibodies both by immunofluorescence and immunoprecipitation studies. TcR/CD3 expression was approximately 10% of normal.
Figure 4. Calcium flux followingTcel1activation. Indo-1-loaded patient (A and C) or control lymphocytes (B and D) were activated either with anti-CD3 mAb (125 ng/ml) plus RaMIg (final dilution 1/100) (A and B) or with anti-CD2 pair [D66 +T11.1; final dilution 1/200; (C and D)]. Samples were monitored discontinuously for 3 min and 30 s in (A) and (B) and for 7 min and 30 s in (C) and (D). Results are expressed as 405/485 nm fluorescence ratio in arbitrary units as a function of time.
Low expression of the T cell receptorlCD3 complex
Eur. J. Immunol. 1991. 21: 1641-1647
1645
Table 2. Lymphocyte proliferationa) Patient
Stimulus
+
m
800f
None (day 4) PHA OKT3 (Snglml) OKT3 (500ng/ml) OKT3 (50ng/d) i-mAb 248 Anti-CD2 ( 0 2 . 1 0 2 . 9 ) PMA (lW9 M) Anti-CD2 (D661+T11.1) PMA ( M)+ ionomycin (lo-* M) None (day 6) Tktanus toxoid
Normal values (cpm x 10-3)
700+ 300 126000f37000 36oOof23oOo 57000f27oOo 225000f35000 110000f 5 5 m 157000f 33000 168ooO+ 62000 800f loo 31000 f 19ooO 34000f21600 35000f 17400
loo0
2300+ 21m+ 16000f
400 600
500 300 9000f 2000
+
1000f
78800+15000 1m+ 300 22500+ 6900 23mf 6400 11500f 2500
candidin Allogeneic cells
a) Results are expressed as mean incorporation of [3H]dThd (cpm 1SD). b) Anti-CD28 antibody (mAb 248) alone did not induce T cell proliferation.
*
Table 3. Expression of activation antigens
P e t c a w of positive
Mean fluorescence
ah
intensity (Iog d e ) Patient Control
.
+
PMA
Patient
Control
76 34 2 8
82 36
cD25 HLA-DR
ionomycin M) OKT3 (SO nglml)
cD25
HLA-DR
487
-
53 40
-
NK activity Patient
50/1
74 67 60 28
2511 131 611
Cl’L activity Normal values
Normal values
Patient
53 k 21
2 2 1
n+9 17 f 7 11 26
ND
ND
41 k 19 97k16 17-r-13
