Eur. J. Immunol. 1990.20: 2051-2056 Absence of Ig rearrangements in SCID 2051

Allan Thompsono, Rudolf W. Hendriksov, Margriet E. M. KraakmanO, ki ts KoningO, Rende Langlois-van den Bergh", Jaak M. Vossenv, Cony M. R. WeemaesO and Ruud K. B. SchuurmanO

Division of Immunobiology and Geneticso, Departments of Immunohaematology and of Pediatricsv, University Medical Center, Leiden and Department of Pediatricso, St. Radboud University Hospital Nijmegen

Severe combined immunodeficiency in man with an absence of immunoglobulin gene rearrangements but normal T cell receptor assembly*

An autosomal recessive type of severe combined immunodeficiency disease (SCID) was characterized by an absence of immunoglobulins (Ig) in the serum and of Ig+ lymphocytes in bone barrow (BM) and peripheral blood. In the BM CDlO+/terminal deoxynucleotidyl transferase-positive lymphocytes were identi- fied. Epstein-Barr virus-transformed B lymphoblastoid cell lines (BLCL) obtained from BM and peripheral blood did not synthesize Ig. The Ig heavy and light chain gene complexes in the BLCL had retained the germ-line configuration. Mature Tcells were present but their numbers in peripheral blood were decreased. T lymphoblastoid cells derived from peripheral blood expressed normal Tcell receptor (TcR) CD3 complexes and manifested various genomicTcR rearrangements. It was concluded that this type of SCID entailed a complete arrest of B lymphocyte differentiation in an early stage prior to Ig rearrangements and a quantitative defect of T lymphocytes which nevertheless allowed develop- ment of matureT cells. Repeated failures of BM transplantation and the striking absence of Ig assembly suggested that this SCID defect resides in the BM microenvironment.

1 Introduction

In man severe combined immunodeficiency (SCID) is not a single disease entitiy, but a heterogeneous group of con- genital disorders characterized by defects in both B and T lymphocytes [ l , 21. Defects in the purine salvage pathway enzymes adenosine deaminase (ADA) or purine nucleoside phosphorylase (PNP) account for about 50% of the auto- soma1 recessive cases [3]. SCID can also result from defective IL2 production, from a defect in the regulation of the expression of MHC determinants or from an abnormal capping of cell surface molecules [ a ] . Apart from these deficiencies SCID can originate from other as yet unknown autosomal defects and from a defect localized on the long arm of the X chromosome [ l , 2, 71.

Most patients have markedly reduced numbers of circulat- ing Tcells with deficient responses to antigens and usually also to mitogens in vitro. Peripheral blood B cell numbers are often normal or increased. In many SCID patients B cells may be functionally intact, suggesting that the failure of B cells to differentiate into Ig-secreting cells originates from the Tcell deficiency. B cells from SCID patients with

[I 84201 * These studies were supported in part by the Dutch Prevention

Fund (28.1607) and the Dutch Foundation for Medical and Biological Research (Medigon 507.113).

Correspondence: Ruud K. B. Schuurman, Dept. of Immuno- haematology E3-Q, University Hospital, Rijnsburgerweg 10, 2333 A A Leiden, The Netherlands

Abbreviations: BLCL: B lymphoblastoid cell line SAC: Sta- phylococcus aureus Cowan I SCID: Severe Combined Immuno- deficiency TdT Terminal deoxynucleotidyl transferase TLC: T lymphoblastoid cells

functional ADA were capable of proliferation to the B cell-specific mitogen Staphylococcus aureus Cowan I (SAC; [S]). In the presence of normal Tcells, SCID B cells from a series of patients exhibited responses to PWM or antigens [9, 101. SCID patients who received BM transplants and manifested engraftment of donor Tcells only, nevertheless started production of antibodies, apparently by the host (SCID) B cells [ l l ] .

However, incidentally, patients have been reported in which B and pre-B cells are completely absent, suggesting a distinct type of SCID with a more profound B cell defect that cannot be explained by the absence of Tcell help only [12,13].This report concerns an autosomal recessive type of SCID characterized by an absence of Ig in the serum and of Ig+ lymphocytes in BM and peripheral blood, whereas mature T cells were present. In order to characterize the molecular basis of the disease, the Ig and TcR gene configuration and expression were investigated in EBV- transformed lymphoblastoid cell lines and T cell clones, respectively.

2 Material and methods

2.1 Patients

The two patients were sisters; the first patient RF was admitted to the hospital with chronic diarrhea and derma- titis. Immunological investigations exposed the underlying SCID. Her sister YF was investigated in the first month of life and was also shown to have SCID. The parents were healthy. Extensive pedigree analysis did not expose other patients with SCID and excluded consanguinity of the parents. Both patients manifested severe malabsorption, growth retardation and failure to thrive. ADA, PNF' and transcobalamine I1 activities were normal. No defects in expression of MHC or the leukocyte surface glycoproteins

0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990 0014-2980/90/0909-2051$3.50 + .25/0

2052 A. Thompson, R. W. Hendriks, M. E. M. Kraakman et al. Eur. J. Immunol. 1990. 20: 2051-2056

LFA-1, Mac-1 and p150,95 were found. The patients were kept in a protective environment; therapy consisted of y-globulin suppletion and if necessary antibiotics. Each patient was grafted three times over a period of 11 months with haploidentical T cell-depleted BM from one of the parents, but never showed any sign of engraftment. RF and YF died from a sepsis and central nervous system compli- cations, respectively, both with BM aplasia.

2.2 B lymphoblastoid cell lines (BLCL)

PBMC isolated by standard Ficoll-Isopaque gradient cen- trifugation and BM cell suspensions were incubated for 1.5 h at 37 “C with EBV-containing SN from cultures of the marmoset cell line B95-8 [ 141. BM cells were cultured in 24-well plates at a concentration of 2.0 x lo6 cells/well and PBMC in 96-well microtiter plates at a concentration of 2.5 x 105 cells/well in RPMI 1640, supplemented with 10% FCS and 5 pg/ml PHA (Wellcome Diagnostics, Dartford, GB). The BLCL were expanded in RPMI 1640 with 10% FCS.

2.3 T lymphoblastoid cells (TLC) and clones

PBMC were cultured at a concentration of 2 x lo5 cells/well in Iscove’s Modified Dulbecco’s Medium (IMDM) supple- mented with 10% human serum, 50 U/ml rIL2 (Cetus Corp. Emeryville, CA), 5 pg/ml pHA and lo6 irradiated (2500 rad) allogeneic PBMC as feeder cells [ 151. Cultures were continued in IMDM containing 10% human serum and 50 U/ml rIL 2. The TLC were further expanded by restimulation with irradiated allogeneic PBMC, 5 pg/ml PHA and 50 U rIL 2 every 8-12 days. T cell clones were established by LD at concentrations of 0.3, 3.0 and 30.0 cells/well in microtiter plates with 105 irradiated allogeneic PBMC as feeder cells.

2.4 Immunofluorescence procedures, analysis of Ig synthesis and lymphoproliferative responses

PBMC and BM cells were analyzed by surface immuno- fluorescence microscopy as described previously [ 161, using murine mAb to CD1, CD3, CD4 and CD8 (Ortho Diag- nostic System Inc., Raritan, NJ), CD5 (Becton Dickinson, Mountain View, CA), CDlO (VIL-Al, W. Knapp, Wien, Austria), CD16 (B73-1, Dr. G . Trinchieri, Philadelphia, PA) and CD19 (Coulter Immunology, Hialeah, FL) and FITC-conjugated goat anti-mouse Ig (Nordic, Tilburg, The Netherlands) as a second antibody.

Cytoplasmic J chain expression was assessed using rhoda- mine-labeled rabbit anti-human J chain (Nordic). Antisera to human Ig H and L chains (Kallestad, Chaska MN and Dakopatts, Glostrup, Denmark) and CD22 (Ly-1, CLB, Amsterdam,The Netherlands) were applied in cytoplasmic as well as surface immunofluorescence techniques. Nuclear terminal deoxynucleotidyl transferase (TdT) was detected by rabbit anti-TdT and FITC-conjugated F(ab’)2 goat anti-rabbit IgG (BRL, Uxbridge, GB).

In the BLCL the fraction of cells secreting Ig was deter- mined by a sandwich ELISA-spot technique in microtiter

wells using goat anti-human IgM, IgD, IgG, IgA, x and h (Tag0 Inc., Burlingame, CA; [17]).

The expression of surface antigens on BLCL or TLC was determined by automated FCM of 1 x 105 - 2 x lo5 cells/assay [ 151. The following mAb were applied: anti-CD9 and anti-CD24 (Hybritech Inc., San Diego, CA); anti- CD1-5, anti-CD7, anti-CD8, anti-CD10, anti-CD16, anti- CD20-22, anti-CD45R, anti-HLA-DR and anti-TcR a/p chain WT31 (Becton Dickinson); anti-IgM, anti-IgD, anti- x , anti-h, anti-CD19 (Dakopatts); anti-TcR y/6 chain IIF2 (J. Borst, Amsterdam,The Netherlands); anti-CD23 (J. De Vries, Dardilly, France) and FITC-conjugated goat anti- mouse Ig (Becton Dickinson) as a second antibody.

Proliferative responses to allogeneic cells, PHA, PWM or antigens (tetanus toxoid, Candida albicans and diphtheria) were determined by measuring incorporation of [3H]dThd 1181.

2.5 Biochemical analyses

BLCL were labeled with [35S]methionine and [3H]lysine [19]. Ig in culture SN and cell lysates were precipitated using SAC and rabbit antisera to Ig H and L chain (CLB). For analysis of TcR complexes T cells were surface iodinated with 1251 and subsequently lysed in 1% digitonin for 20 min [15]. After removal of insoluble material by centrifugation (13000 x g) for 30 min the lysates were incubated with protein A agarose beads (Pharmacia, Uppsala, Sweden) and normal rabbit serum. Immunoprecipitation was per- formed using an anti-CD3 mAb (OKT3, Ortho; [15]).

The samples were analyzed by SDS-PAGE as described previously [ 191. Gels with [35S]methionine- and [3H]lysine- labeled samples were subjected to fluorography using 2,5-diphenyloxazole (PPO, Serva, Heidelberg, FRG) and exposed to X-0-mat R film (Eastman Kodak Co. Roches- ter, NY).

2.6 Genomic DNA analyses

High molecular weight genomic DNA was isolated from BLCL and TLC using standard procedures. DNA (10 pg) was digested with the appropriate restriction endonu- cleases (Promega Biotech, Madison, WI) and further processed in Southern blotting using GeneScreenPlus hybridization transfer membrane (Dupont Company, Bos- ton, MA; [20]).

The following DNA probes were applied: JH region (2.5-kb Eco RI-Bgl I1 fragment) and CH region (1.3-kb Eco RI fragment; [21]); J, region (2.0-kb Sac I fragment; [22]); Ca region (3.5-kb Eco RI-Hind I11 fragment; [23]); TcR Cg region (JUR-beta-2 Bgl I1 fragment; [24]). Probes were labeled by random priming (Boehringer Mannheim, Mann- heim FRG).

3 Results

3.1 Characterization of the SCID

The sistersYFand RF had a leukopenia and a lymphopenia (1.4 x lo9 - 2.6 x lo9 IymphocytesAiter) which was

Eur. J. Immunol. 1990. 20: 2051-2056 Absence of Ig rearrangements in SCID 2053

progressive (0.6 x lo9 - 0.9 x lo9 lymphocytes/liter for YF and RF at 2 years of age). IgM, IgA, IgD or IgE were not detectable in the sera using nephelometric and ultralow immunodiffusion techniques. The levels of maternally derived IgG decreased to < 0.2 mg/ml in RF at 12 months and to 1.6 mg/ml inYFat 5 months at which ages treatment with i.v. Ig was started.

In peripheral blood no sIg+ cells were detected by immu- nofluorescence microscopy with antisera to Ig Cp or C6 region or to x. or h L chain (Table 1). No lymphocytes that expressed other B or pre-B cell-specific determinants were found (CD10, CD19 or CD22,Table 1). Cells committed to the B cell lineage could be identified in the BM: about 80% (normal range: 2.5%-12%, [16]) of the lymphoid cell expressed (TdT and the common acute lymphoblastic leukemia antigen CDlO (Table 1). No sIgM+ B lympho- cytes or pre-B lymphocytes that expressed Ig p chain in the cytoplasm were found (normal values 20 .9%4.0% and 10.7%-25.4% of lymphocytes, respectively [16]). Ig H or L chain-containing plasma cells were absent and no lymphoid cell expressed CD22 in the cytoplasm (Table 1).

In normal BM 80% of TdT+ lymphocytes contain CD22 and 5% also have Ig p chain in the cytoplasm [25].Therefore, in this SCID the B lymphocyte disorder entails an arrest in early B cell differentiation at a stage whereTdT is expressed but before the expression of CD22 or Ig p chains.

Initially the percentages of CD3+ cells in peripheral blood were almost normal and were 30%-40% at the age of 2-3 years (Table 1). Because of a concomitant progressive lymphopenia the absolute numbers of CD3+ T lympho-

Table 1. Immunofluorescence analysis

RF YF (% of lymphoid cells)

Peripheral blood m C,a) m Cb m X , h m CD19 m CD22 m CDlO CD 1 CD3 CD4 CD8 CD16 BM CY c, m c,, Nuclear TdT CDlO cy CD22 CDl CD3 CD4 CD8 CD5

a) cy = cytoplasmic; m = membrane.

0 0 0 0 0 ND 0 0 0 0 0 0 0 0 40 30 42 12 6 16 44 66

0 0 0 0 a4 84 74 78 0 0 0 1 3 3 3 1 2 2 1 1

cytes in peripheral blood decreased to < 10% of normal. These T lymphocytes carried the CD4 or CD8 determi- nants. No Tcells of an immature phenotype (CDl+) were present in peripheral blood (Table 1). A substantial pro- portion of the peripheral blood lymphoid cells expressed the FCyR CD16, indicative of NK cells (Table 1).

In v i m proliferative responses to PHA decreased from 60% to 10% and to PWM from 90% to 25% of the responses of PBMC of healthy blood donors. PBMC of RE following the regular vaccinations, had positive prolifera- tive responses to Candida albicans, tetanus toxoid and diphtheria that were within normal range. At 1 month of age PBMC of YF had normal allogeneic responses in MLC, while PBMC of RF at 1 year of age yielded responses that corresponded to 10-20% of the reactivity of PBMC of healthy blood donors.

3.2 Characteristics of EBV-transformed lymphoblastoid cell Lines

Lymphoblastoid cell lines were obtained from BM cell suspensions (RF.BM and YF.BM) and PBMC (YEPB 1-8) by transformation with EBV. In FCM analysis these cell lines were negative for the Tcell determinants CD1, CD3, CD4, CD5, CD7 and CD8, but expressed the B cell determinants CD19, CD20, the C3dEBV receptor CD21, CD22, the FceR CD23, CD24, as well as CD45R and HLA-DR. Expression of nuclear TdT and pre-B cell determinants CD9 or CDlO was negative. As this marker phenotype was identical to that of BLCL derived from PBMC of healthy controls [26] it was concluded that these lymphoblastoid cell lines were derived from lymphocytes committed to the B cell lineage.

All BLCL of both patients were found to lack secretion of Ig H or L chain when analyzed by a sandwich ELISA-spot assay. The BLCL were also negative for cytoplasmic Ig in immunofluorescence microscopy or for sIg by FCM. All BLCL expressed the 15-kDa J chain protein as determined by cytoplasmic fluorescence.

No Ig-specific bands were exposed in SDS-PAGE analysis after immunoprecipitations from total cell lysates or culture SN of all BLCL, neither with SAC only, nor with antisera to all Ig H or L chain isotypes.

3.3 Ig H and L chain gene configuration in the BLCL

DNA from the BLCL was digested with Hind 111, Bgl 11, Barn HI, Eco RI and Xba I and after Southern blotting assayed for hybridization with a human JH probe. Only germ-line bands identical in size to those found in liver DNA were detected (shown for Bgl I1 and Eco RI in Fig. 1A and B). Therefore, in all BLCL both Ig H chain alleles had retained the germ-line configuration. When C, probe was utilized again only germ-line bands were detected, which excluded recombinations in the JH-C,, intron region. When the same digests were hybridized with an J, or a CA probe no rearrangements were detected (shown for J, in Bgl I1 and Eco RI digests in Fig. 1C and D). In all BLCL both alleles of the Ig x and h L chain loci had retained the germ-line configuration.

2054 A. Thompson, R. W. Hendriks, M. E. M. Kraakman et al. Eur. J. Immunol. 1990. 20: 2051-2056

Figure 2. FCM analysis of RV TLC and Y E S . and YF.sT2 Tcell clones. Cells were labeled with fluorescein-conjugated goat anti-mouse Ig antibodies after incubation with medium only (-) or mAb specific for TcR cdp or CD3. Ordinate: cell number; abscissa: log fluorescence intensity.

Figure 1 . Southern blot analysis. Genomic DNA from human liver and the indicated RF and YF BLCL was digested with (A) Bgl I1 and (B) Eco RI and hybridized to a JH segment probe. Digests with (C) Hind 111 and (D) Eco RI were hybridized to a J, segment probe. The position of the germ-line DNA fragment is indicated by arrows and the sizes are in kb.

3.4 Analysis of TcR expression by TLC

From peripheral blood TLC were obtained by PHA/IL 2 stimulation. By subsequent LD individual T cell clones from the YF TLC were established, designated YF.sT1, YF.sT2 and YF.sT5. By FCM analysis theTLC from RF and YF Tcell clones were shown to expressTcR a and p chains and the CD3 determinant (Fig. 2). No reactivity with the TcR $&specific antibody l lF2 was observed. The TLC expressed theTcell determinants CD2 and CD7.TheYF T cell clones had the CD4+CD8- phenotype, whereas the RF TLC were found predominantly CD4+ but also CD8+.

YF T cell clones were surface labeled and immunoprecip- itated with an anti-CD3 antiserum. Subsequent SDS- PAGE analysis exposed TcR complexes with apparent molecular masses of 70-90 kDa. These complexes were reduced to 40-50-kDa subunits that appeared to be unique for the individual T cell clones (Fig. 3).

Figure 3. SDS-PAGE analysis of anti-CD3 immunoprecipitates from digitonin lysates of lz5I cell surface-labeled YF Tcell clones under non-reducing (left) or reducing conditions (right). The positions of the TcR and CD3 molecules are indicated. Molecular weight markers are in kDa.

In Southern blotting experiments Bam HI digests of DNA from the RF TLC and theYFsTclones were hybridized with aTcR Cg probe (Fig. 4). Apart from a 24-kb germ-line band also present in liver DNA, every YF Tcell clone showed a different rearranged band. In the RF TLC one predomi- nant and several other rearranged bands were detected. Thus the TLC from both patients had various genomicTcR rearrangements and expressed intact TcR a@ - CD3 complexes.

4 Discussion

The occurrence of an identical immunodeficiency in two female siblings from non-consanguinous healthy parents

Eur. J. Immunol. 1990. 20: 2051-2056 Absence of Ig rearrangements in SCID 2055

240

Figure 4. Southern blot analysis. Genomic DNA from human liver,YF Tcell clones and RF TLC was digested with Barn HI, Eco RI and hybridized to a human TcR Cp probe. The arrow indicates the position of the germ-line DNA fragment (size in kb).

supported an autosomal recessive inheritance of a single genetic defect. The combination of a lack of antibody production and decreased T lymphocyte numbers estab- lished the diagnosis SCID.The well-characterized defects in SCID, i.e. deficiency of ADA, PNP or h4HC expression were excluded.

In the BM no sIg+ B cells or cytoplasmic Ig p chain-positive pre-B cells were present. Only TdT+/CDlO+ lymphocytes that did not express cytoplasmic CD22 were found. AsTdT is supposed to have a function in the Ig gene rearrange- ments [27], these lymphocytes most likely reflect a stage at which the Ig V(D)J recombination is expected to start.

Despite the lack of Ig expression B lymphoblastoid cell lines could be established by EBV transformation of BM and PBL. These BLCL expressed B cell surface determi- nants of a mature B cell phenotype, which are normally found on Ig-producing BLCL [26]. Extensive immunolog- ical and biochemical investigation of the SCID BLCL did not expose any Ig production. The SCID BLCL had retained the germ-line configuration on both alleles of the Ig H and L chain complexes.

The T lymphocyte deficiency was characterized by a reduction of peripheral blood T lymphocyte numbers: Tcell numbers gradually decreased to < 5% of normal. As a consequence the responses upon stimulation by lectins and allogeneic cells were also decreasing. Nevertheless, the T lymphocytes responded to microbial antigens. TLC derived from peripheral blood expressed normal TcR-CD3 com- plexes and manifested various genomic TcR rearrange- ments. In summary, the SCID reported here entails (a) a complete arrest in an early stage of B lymphocyte differ- entiation prior to the initiation of Ig rearrangements and (b) a quantitative defect in the T cell lineage which

nevertheless allowed development of mature T lympho- cytes.

It might be postulated that the SCID mutation affects the V(D)J recombination process itself in the B cell lineage only. However, a defect in one of the components of the recombinase system seems unlikely as all the presently identified factors have been implicated in both B and Tcell rearrangements: the RAG-1 gene which can activate recombination in 3T3 fibroblasts is expressed in pre-B cells as well as immatureTcells [28], the conserved heptamer and nonamer recombination signal (RS) sequences recognized by the RS-binding protein [29] flank Ig as well as TcR gene segments [30, 311. In addition, Ig D to J rearrangements have been found inT lymphocytes [32] and Abelson murine leukemia virus-transformed pre-B cell lines frequently rearrange the TcR y locus [33].

The described human SCID mutation appears to be different from the recombinase defect in the CB-17 SCID mice [34,35] which leads to an impairment of bothTcR and Ig rearrangements, probably becauseV(D)J gene segments cannot be correctly joined [36,37]. Although it is concluded that a defect in the recombinase itself in this human SCID is unlikely, the striking absence of Ig rearrangements may suggest a defect in a B cell-restricted activator of the recombinase system. This SCID mutation should then involve a pleiotropic factor that is crucial for Ig recombi- nation in B cells and important for the generation of appropriate Tcell numbers. The pleiotropic nature of such a factor is not surprising, as the regulators of B cell differen- tiation identified so far all have a wide range of activities [38]. The repeated failures in transplantation of allogeneic BM cells suggest that this pleiotropic factor is produced by BM stroma cells as the BM microenvironment is known to be critical for early B cell development [39] and less so for T cell maturation as TcR assembly occurs in the thymus.

Received March 23, 1990.

5 References

1 Gelfand, E.W. and Dosch, H.-M., in Wedgwood, R. J., Rosen, F. S. and Paul, N. W. (Eds.), Primary immunodeficiency diseases, Birth Defects: Original Articles Series,Vol. 19, Alan R. Liss Inc., New York 1983, p. 65.

2 Rosen, F. S., Cooper, M. D., and Wedgwood, R. J. €?, N. Engl. J. Med. 1984. 311: 300.

3 Hirschhorn, R., Clin. Immunol. Immunopathol. 1986. 40: 157.

4 Pahwa, R., Chatila, T., Pawha, S., Paradise, C., Day, N. K., Geha, R., Schwartz, S. A., Slade, H., Oyaizu, N. and Good, R. Proc. Natl. Acad. Sci. USA 1989. 86: 5069.

5 Schuurman, R. K. B., Van Rood, J. J., Vossen, J. M., Schellekens, €? Th. A., Feltkamp-Vroom,Th. M., Doyer, E., Gmelig-Meyling, F. and Visser, H. K. A., Clin. Immunol. Immunopathol. 1979. 14: 418.

6 Gelfand, E.W., Oliver, J. M., Schuurman, R. K. B., Matheson, D. S. and Dosch, H.-M., N . Engl. J. Med. 1979. 301: 1245.

7 De Saint Basile, G., Arveiler, B., Oberlt, I., Malcolm, S., Levinsky, R. J., Lau,Y. L., Hofker, M., Debrk, M., Fischer, A., Griscelli, C. and Mandel, J. L., Proc. Natl. Acad. Sci. USA 1987. 84: 7576.

8 Schuurman, R. K. B., Gelfand, E. W. and Dosch, H.-M. J. Immunol. 1980. 125: 62.

2056 A. Thompson, R. W. Hendriks, M. E. M. Kraakman et al. Eur. J. Immunol. 1990. 20: 2051-2056

9 Pahwa, S. G., Pahwa, R. N. and Good, R. A., J. Clin. Invest. 1980. 66: 543.

10 Gelfand, E. W. and Dosch, H.-M., Clin. Immunol. Immuno- pathol. 1982. 25: 303.

11 Buckley, R. H., Schiff, S. E., Sampson, H. A., Schiff, R. I., Markert, M. L., Knutsen, A. P., Hershfield, M. S., Huang, A. T., Mickey, G. H. and Ward, F. E., J. Immunol. 1986. 136: 2398.

12 Hayward, A. R., Lancet 1978. i: 1014. 13 Ichihara,Y., Matsuoka, H.,Tsuge, I., Okada, J.,Torii, S.,Yasui,

H. and Kurosawa,Y., Immunogenetics 1988. 27: 330. 14 Miller, G. and Lipman, M., Proc. Natl. Acad. Sci. USA 1975.

70: 190. 15 Koning, F., Knot, M. ,Wassenaar, F. and Van den Elsen, P., Eur.

J. Immunol. 1989. 19: 2099. 16 Asma, G. E. M., Langlois van den Bergh, R. and Vossen, J. M.,

Transplantation 1987. 43: 865. 17 Hendriks, R.W.,Thompson, A., Kraakman, M. E. M., Drayer,

A. L. andschuurman, R. K. B., Clin. Exp. Immunol. 1988.74: 305.

18 Schot, J. D. L., Elferink, D., Hooijkaas, H., Neijens, H. J. and Schuurman, R. K. B., Clin. Exp. Immunol. 1987. 68: 48.

19 Scholten, P., Schuurman, R. K. B. and Ploegh, H., Hum. lmmunol. 1986. 16: 1.

20 Hendriks, R. W.,Van Tol, M. J. D., De Lange, G. G. and Schuurman, R. K. B., Scand. J. Immunol. 1989. 29: 535.

21 Ravetch, J.V., Siebenlist, U., Korsmeyer, S.,Waldmann,T. and Leder, P., Cell 1982. 27: 583.

22 Hieter, P. A., Maizel, J.V. and Leder, P., J. Biol. Chem. 1982. 257: 1516.

23 Hieter, P. A., Hollis, G. F., Korsmeyer, S. J. ,Waldmann,T. A. and Leder, P., Nature 1981. 294: 536.

24 Yoshikai,Y., Anatoniou, D., Clark, S. P. ,Yanagi,Y., Sangster, R. ,Van den Elsen, I? ,Terhorst, C. and Mak,T. W., Nature 1984. 312: 521.

25 Campana, D., Janossy, G., Bofill, M.,Trejdosiewicz, L. K., Ma, D., Hoffbrand, A.V., Mason, D.Y., Lebacq, A.-M. and Forster, H. K., J. Imrnunol. 1985. 134: 1524.

26 Ling, N. R., in Bird, G. and Calvert, J. E. (Ed.), B Lympho- cytes in human disease, Oxford University Press, Oxford 1988, p. 174.

27 Landau, N. R., Schatz, D. G., Rosa, M. and Baltimore, D., Mol. Cell. Biol. 1987. 7: 3237.

28 Schatz, D. G., Oettinger, M. A. and Baltimore, D., Cell 1989. 59: 1035.

29 Matsunami, N., Hamaguchi, Y., Yamamoto, Y, Kuze, K., Kangawa, K., Matsuo, H., Kawaichi, M. and Honjo,T., Nature 1989. 342: 934.

30 Tonegawa, S., Nature 1983. 302: 575. 31 Kronenberg, M., Sui, G., Hood, L. and Shastri, N., Annu. Rev.

Immunol. 1986. 4: 529. 32 Kurosawa, Y., Von Boehmer, H., Haas, W., Sakano, H.,

Traunecker, A. and Tonegawa, S., Nature 1981. 290: 565. 33 McCubrey, J. A., Steelman, L. S., Risser, R. G. and McKearn,

J. I?, Eur. J. Immunol. 1989. 19: 2303. 34 Bosma, G. C., Custer, R. P. andBosma, M. J., Nature 1983.301:

527. 35 Schuler,W.,Weiler, I. J., Schuler, A., Philips, R. A., Rosenberg,

N., Mak,T.W., Kearney, J. F., Perry, R. P. and Bosma, M., Cell 1986. 46: 963.

36 Hendrickson, E. A., Schatz, D. G. and Weaver, D. T., Gene Devel. 1988. 2: 817.

37 Malynn, B. A., Blackwel1,T. K., Fulop, G. M., Rathbun, G., Furley, A. J. W., Femer, P., Heinke, L. B., Phillips, R. A., Yancopoulos, G. D. and Alt, F. W., Cell 1988. 54: 453.

38 OGarra, Umland, S., De France, T. and Christiansen, J., Immunol. Today 1988. 9: 45.

39 Whitlock, C. A. and Witte, 0. N., Proc. Natl. Acad. Sci. USA 1982. 79: 3608.