Efficient isolation of mutant antigen presenting cell lines by functional selection using T cell clones

JOURNALOF INIMuwou)6;lCAL METHODS

ELSEVIER

Journal

of immunological

Methods

182 (1995) 209-218

Efficient isolation of mutant antigen presenting cell lines by functional selection using T cell clones ” Syuichi Koarada a, Eiroh Kubota b, Miyoko Tokushima a, Keiko Naitoh a, Kensuke Miyake a, Masao Kimoto a,* a Department of immunology, Saga Medical School, Nabeshima, Saga 849, Japan h Department of Oral Surgery, Saga Medical School, Nabeshima, Saga 849, Japan Received

22 December

1994; accepted

8 February

1995

Abstract An efficient method for the isolation of mutant antigen-presenting cell (APC) lines is described. When mixtures of transfectant APC lines TApz (that express A/?z/Aad MHC class II molecules) and hypothetical variant APC lines TAPd (that express A@d/ALud class II molecules) were cultured with and selected by autoreactive Apz/Aad-restricted T cell clones, the percentage of TAPd APC lines increased from less than 1% of the original APC mixtures to almost 100% after several cycles of selection. This increase of hypothetical variant was shown to be due to the formation of aggregates of wild-type TApz APC lines with Apz/Aad-restricted autoreactive T cell clones that results in the inhibition of proliferation and probably killing of TApz APC lines. Based on this, ethyl methane sulfonate (EMS)-treated TAPz APC lines or B-B hybridoma APC lines MW4 (that express Apz/Aad and Apz/Aaz class II molecules) were cultured with and selected by A@z/Aad-restricted autoreactive T cell clones to obtain mutant APC lines that escaped the recognition by T cell clones. After cloning, about 43% of clones examined lost the ability to stimulate T cell clones with concomitant loss of class II molecule expression. Less than 1% showed loss of stimulatory activity against T cell clones in spite of the expression of normal amounts of class II molecules. Initial analysis revealed that they include APC mutant lines with (1) altered MHC class II sequences, (2) loss of adhesion molecule expression and (3) possible impairment of the peptide loading. The method described here may provide a variety of mutant APC lines that are useful for the analysis of antigen processing and presentation pathways as well as of class II structure for T cell stimulation. Keywords:

Antigen presenting

cell; T cell clone; Mutant; Major histocompatibility

complex class II molecule -

1. Introduction Abbreviations: MHC, major histocompatibility complex; APC, antigen presenting cell; EMS, ethyl methane sulfonate; KLH, keyhole limpet hemocyanine. ‘This work was supported in part by grants from the Ministry of Education, Science, and Culture of Japan. * Corresponding author. Tel.: (0952) 31-6511; Fax: (0952) 33-2518. 0022-1759/95/$09,50 0 1995 Elsevier SSDI 0022.1759(95)00051-8

Science

Antigen-specific receptors on CD4+ T lymphocytes recognize antigenic peptides bound within the groove of major histocompatibility complex (MHC) class II molecules (Germain,

B.V. All rights reserved

210

S. Koarada et al. /Journal

of Immunological Methods 182 (1995) 209-218

1994). These antigenic peptides are shown to be created by partial proteolytic digestion of endocytosed foreign materials or of endogenous self-antigens and are believed to be trapped in the endosome of an acidic environment where invariant chains detach from class II molecules to make a room for these peptides (Cresswell, 1994). Class II molecules with antigenic peptides are then expressed on the surface of antigen-presenting cells @PCs) and recognized by CD4+ T lymphocytes. However, the precise mechanism for the processing and presentation of such antigenic peptides is still unknown. Mutant APC lines have been shown to be a powerful tool for the analysis of antigen processing and presentation to T cells (Germain and Malissen, 1986; Glimcher and Griffith, 1987). These mutant APC lines were created by transfection of mutated class II genes or by immunoselection of chemically mutagenized APC lines. Another powerful tool for the isolation of APC mutant lines is functional screening using cloned antigen-specific T cell hybridomas as developed by Dang et al. (Dang et al., 1990,1993). The strategy of this functional screening, in theory, enables us to isolate APC mutants having defects anywhere in the entire processing and presentation pathways of antigenic peptides/class II complex, as well as alterations in class II sequences. In this paper, we describe an efficient method for the isolation of various mutant APC lines by functional screening using antigen-specific T cell clones. We also describe the partial characterization of these mutant APC lines. The results suggest the potential usefulness of this procedure to isolate a variety of mutant APC lines that may provide unique materials for the analysis of antigen processing and presentation as well as of the class II structure. 2. Materials and methods 2.1. Mice New Zealand Black (NZB, H-2d) and New Zealand White (NZW, H-2”) mice were purchased from Japan SLC (Shizuoka, Japan). (NZB X NZWIFl (B/WFl) mice were made by mating

female NZB with male NZW mice in our animal breeding facilities. Animals in this study were maintained and used according to the Saga Medical School Guidelines for Animal Experimentation. 2.2. Monoclonal antibodies (rnAbs) The origins and specificities of hybridoma cell lines 10.2.16 (anti-A/3z> (Oi et al., 19781, K24-199 (anti-Aad) (Koch et al., 1982), 4D5-12 (anti-Acuz) (Beck et al., 19861, MKD6 (anti-Apz> (Kappler et al., 19811, YN1./1.7 (anti-IQ&I-l) (Takei, 19851, and FD441.8 (anti-LFA-1) (Sarmiento et al., 1982) are described in each reference. A rat anti-mouse monomorphic I-A mAb (BW/9) was made in our laboratory (Yamashita and Miyake, unpublished). Our preliminary immunofluorescence staining experiments revealed that 10.2.16 and 4D5-12 mAbs react to NZW (H-2”) but not to NZB (H-2d) spleen cells. K24-199 mAb reacts to NZB but not to NZW spleen cells. Cultured supernatants of hybridoma cell lines or appropriately diluted ascitic fluids were used for immunofluorescence staining. 2.3. T cell clones The method for the establishment of T cell lines and clones has been described previously (Gotoh et al., 1993; Tokushima et al., 1994). Keyhole limpet hemocyanine (KLH) was purchased from Calbiochem-Behring Corp. (La Jolla, CA>. Complete medium consists of RPMI-1640 culture medium (Gibco, Grand Island, NY) containing 10% heat-inactivated fetal calf serum (FCS) (Intergen, Purchase, NY), 5 X 10e5 M 2mercaptoethanol, 10 mM Hepes (Gibco), 100 U/ml penicillin and 100 pg/ml streptomycin. r_-glutamine was added at a final concentration of 2 x low5 M before use. Repeated limiting dilution cloning was used to isolate clones of interest. Proliferative responses of T cell clones were assayed as described (Gotoh et al., 1993; Tokushima et al., 1994). Thus, 1 X lo4 T cell clones and 5 x lo3 to 5 x lo4 transfectant cells or B-B hybridomas were cultured in 0.2 ml complete culture medium. After 24 h, 0.1 ml supernatant were transferred to wells containing 1 X lo4 cytokine-

S. Koarada et al. /Journal of Immunological Methods 182 (1995) 209-218

dependent CT-6 cells (Stadler et al., 1981) in 0.1 ml complete medium. After 48 h, proliferative responses of CT-6 cells were assayed by adding 0.5 PCi L3HlTdR for the final 16 h and the uptake of L3H]TdR was measured on a Betaplate flat-bed liquid scintillation counter (PharmaciaWallac, Gaithersburg, MD). We used autoreactive A@z/Acud-restricted T cell clones (KGU108, KGUllS, KGU127 and KGU140) derived from B/WFl mice (Tokushima et al., 1994). 2.4. Transfectant and B-B hybridoma antigen-presen ting cells A M12.C3 cell line, a variant of immunoselected I-A negative mutant of BALB/c (H-2d)derived B lymphoma M12.4.1, was established by Glimcher et al. (1985) and obtained from D. McKean (Mayo Clinic, Rochester, MN). Antigen-presenting transfectant cell lines TA/?z and TA/3d, that express Apz/Aad and Apd/Aad class II molecules respectively, were prepared by transfecting pCEXVAPz (kindly provided by H. Nishimura, Yokohama University) or pCEXVAPd (kindly provided by J. Miyazaki, Tohoku University) into M12.C3 cells as described (Gotoh et al., 1993). A stable B-B hybridoma cell line MW4 was obtained by fusing M12C3 with splenic B cells from NZW mice. MW4 cells express ABz/Aaz and A@z/Aad molecules on their surface with antigen-presenting activities as described (Gotoh et al., 1993). 2.5. Isolation of mutant APC lines This was performed as described by Dang et al. (1990) with modifications. 5 x lo6 TA/3z transfectant APC cell lines or MW4 B-B hybridoma APC cell lines were cultured with 10 ~1 ethyl methane sulfonate (EMS) (Sigma, St. Louis, MD) in 25 ml culture medium for 12 h. The cells were harvested, washed three times to remove EMS, and cultured for several days to allow mutations to be expressed. Cells were harvested and 5 x lo4 EMS-treated APC lines were cultured with 5 x 10’ autoreactive Apz/Aad-specific T cell clones in 2 ml culture medium in a 24-well plate (Falcon No. 3047, Becton Dickinson Lab-

211

ware, Franklin Lakes, NJ). During this culture, T cell clones made aggregates with wild-type APCs, but not with mutant APCs. After 24 h, cultures were gently suspended by a Pasteur pipette so as not to disrupt the T cell clone/APC aggregates. Cells were incubated for a further 10 min, during which time large aggregates settled down while APC lines that did not make aggregates were still in suspension. T cell clones that did not make aggregates also exist in suspension. About 1 ml from the upper half of the culture were transferred to new 24-wells and cultured for several days. During this second culture, wild-type APCs that made aggregates with T cell clones died for unknown reasons (Dang et al., 1990), while mutant APCs that did not react with T cell clones proliferated. T cell clones gradually decreased in number during this second culture. Mutant APC lines were subjected to limiting dilution cloning and screened for their stimulatory activity against various T cell clones and for their reactivity against a panel of anti-class II mAbs and anti-adhesion molecule mAbs. 2.6. Flow cytometry analysis 1 x lo6 APC lines were first incubated with 20 ~1 of appropriately diluted rnAb for 30 min at 4°C. Cells were washed three times with cold PBS (pH 7.2) containing 1% FCS and 0.1% NaN, (staining buffer). Then 20 ~1 of appropriately diluted FITC-protein A (Zymed Lab., San Francisco, CA) or FITC-MAR18.5 (Lanier et al., 1982) was added and incubated for 30 min at 4” C. This was followed by addition of 20 ~1 of 10 pg/ml propidium iodide (Sigma, St. Louis, MO) for the final 5 min. Cells were then washed three times with cold staining buffer and analyzed by FACScan (Becton-Dickinson Immunocytometry Systems, Mountain View, CA).

3. Results 3.1. SelectiL’e enrichment of hypothetical Llariant APC lines by co-cutture with T cell clones Dang et al. (1990) reported that A20 B lymphoma APC lines first make aggregates with T

S. Koarada et al. /Journal

212

of Immunological Methods 182 (1995) 209-218

cell hybridomas of appropriate specificity and then die for unknown reasons. By taking advantage of this, they succeeded in isolating chemically induced mutant APC variant lines that do not react with T cell hybridomas. Accordingly, we studied whether it is possible to isolate variant APC lines by co-culture of T cell clones and APC lines. To learn the feasibility of this procedure, we used the TAPd APC line that expresses A/?d/Aad class II molecules as a hypothetical variant APC line, because this TApd APC line does not make aggregates with autoreactive T cell clone KGU108. We mixed TApz and TAPd APC lines at various ratios and then cultured them with KGU108 T cell clones for 24 h. Cells were suspended by Pasteur pipettes, then allowed to settle down for 10 min. The upper half of the culture that might contain mainly non-aggregated APCs was transferred to new wells. After culture for several days, cells were stained with anti-class II mAbs. As shown in Fig. 1, the percentage of TAPz cells generally increased after selection by KGU108 T cell clones. It should be noted that

50%

16%

from less than 1% in the original APC mixtures, TAPd lines showed an expansion to 1.8% of the APC mixture after co-culture with KGU108 T cell clones. After the second cycle of selection by KGU108, this TApd APC line increased to 5.4% of the final APC mixtures. Repeated selection by co-culture with KGU108 resulted in enrichment of hypothetical variant TApd lines from less than 1% to almost 100% (data not shown). The result suggests that this procedure is a very powerful method for the enrichment of variant APC clones that have escaped from T cell recognition. 3.2. Growth of APC lines is inhibited by contact with T cell clones One of the bases for the successful isolation of rare variants observed in the above experiment could be that growth of APC lines with appropriate specificity was inhibited by contact with T cell clones. To examine this, TApz (as a wild-type APC line) and TAPd (as a hypothetical variant APC line) cells were cultured with varying num-

3.0%

0.8%

0.1%

1

1

TApd ratio 1

1

1

1st Selction

2nd Selection

Fluorescence Intensity

Fig. 1.2 x lo5 TA/+z and TApd class II-transfectant cells at various ratios were cultured with 1 X lo6 auto-reactive A/3z/Arud-restricted T cell clone KGUlO8 in 2 ml culture medium. After 24 h, cultures were gently suspended, settled for 10 mm and the upper halves were transferred to the new wells as described in the text. After several days when the cell mixtures proliferated to saturation, transfectant cells were stained with anti-Afid mAb (MKD6) and analysed by FACScan (1st selection). Residual T cell clones were gated out from the analysis by forward and side scatter. After the 1st selection, the mixture of transfectant cells was cultured again with the KGU108 T cell clone and analysed in a similar way (2nd Selection). Percent positive staining with anti-Apd mAb (MKD6) is indicated in each panel.

S. Koarada et al. /Journal

.OOOl

,001

mAb 10.2.16

.Ol

of Immunological Methods 182 (1995) 209-218

.l

1

(mg/ml)

Fig. 2. 2X lo5 TAPz or TAPd class II-transfectant cells were cultured with 1 X106 irradiated (3300 rad) auto-reactive Apz/Aad-restricted KGU108 T cell clones for 24 h. Varying amounts of anti-Apz mAb (10.2.16) were included at the initiation of culture. Cultures were pulsed with 13H]TdR for the final 4 h.

bers of irradiated (3300 rad) KGU108 T cell clones for 24 h and the proliferative responses of APC lines were assayed by the uptake of [3H]TdR. As shown in Table 1, the proliferative responses of TApz cells were inhibited by co-culture with KGU108 T cell clones in a dose-dependent manner. In contrast, TAPd APC lines that were not recognized by KGU108 did not show any decrease in proliferative response by co-culture with any T cell clones. The result suggested that the growth inhibition of TA/?z cells is mediated by the contact of specific TCR on KGU108 cells with A/3z/Aad class II molecules. To confirm this, anti-Apz class II mAb (10.2.16) was included in this culture. As shown in Fig. 2,

213

anti-Apz mAb permitted the inhibition of TApz APC growth by KGU108 in a dose-dependent manner. The proliferation of TAPd APC lines was always low compared to that of TAPz APC lines, probably due to the intrinsic nature of this cell line. Simple mixed cultures of TApz and TAPd cells resulted in the overgrowth of TApz cells, which may reflect the slow growth of TAPd cells as compared to TApz cells (data not shown). We were not able to demonstrate killer activity of KGUlO8 T cell clones against TApz cells by a conventional 4 h and 12 h killer assay (data not shown). Because the number of TApz APC lines decreases to almost zero after several cycles of selection (see above), TApz cells may die by contact with KGU108 due to unknown mechanisms, as reported by others (Dang et al., 1990). 3.3. Isolation of mutant APC lines by co-culture with T cell clones

We then tried to isolate mutant APC lines from TApz or MW4 cells (created by fusion of M12.C3 B lymphoma APC lines with NZW spleen B cells) by co-culture with autoreactive A/?z/Aad-restricted T cell clones derived from B/WFl mice. The reason for using these autoreactive T cell clones for the isolation of mutant APC lines is that some of these clones (KGU108 and KGU140) showed pathogenicity upon transfer to pre-autoimmune B/WFl mice (Tokushima et al., 1994), so that the mutant APC lines which escaped recognition by these pathogenic T cell clones, would be useful for the analysis of pathogenic self-peptide processing and presentation. TApz or MW4 cells were chemically mutag-

Table 1 Inhibition of proliferation of transfectant APCs by T cell clones APC TApz TApd

[3]HTdR uptake (cpm + SD) no. of irradiated T cell clones (-)

1 x 103

1 x lo4

1 x 105

16861 f 2333 3811 f 453

13 886 + 2 243 3 075 + 303

7 751 f 689 3361 f 214

3 450 f 225 4 938 f 329

2 x 10’ TApz or TAPd class II-transfectant cells were cultured with varying numbers of irradiated (3300 rad) auto-reactive Apz/Aad-restricted T cell clones for 24 h. Proliferative responses of transfectant cells were assayed by the uptake of 1 PCi of [3H]TdR for the final 4 h.

S. Koarada et al. /Journal of Immunological Methods 182 (1995) 209-218

214

enized using EMS and selected by T cell clones as described in the materials and methods section. Growing clones from limiting dilution cultures were examined for their stimulatory activity against KGU series clones and the expression of class II molecules. Table 2 shows 12 independent experiments for the isolation of EMS-treated mutant APC lines. As shown, the efficiency of mutant isolation varies from one experiment to another from 100% to 0%. In a total of more than 1700 clones, about 43% of mutant APC lines lost the ability to stimulate KGU clones. Almost all of these non-functional APC mutant lines concomitantly lost the expression of class II molecules. Less than 1% of the clones showed loss of APC activity, in spite of the expression of normal amounts of class II molecules. We performed partial characterization of representative mutant APC lines for their antigen-presenting activity against a KLH-reactive Apz/Aad-restricted T cell clone and autoreactive Apz/Aad-restricted T cell clones. The expression of surface class II molecules using allele-specific and monomorphic mAbs and the expression of adhesion molecules (LFA-1 and ICAM-1) were also analyzed.

3.4. Mutant APC lines that lost I-A molecule expression As shown in Figs. 3A and 3B, mutant APC lines 7Cll from MW4 cells and 2A10, 8E6, 2C8, 2C3 and 13FlO from TApz cells lost the APC function concomitantly with the loss of class II molecule expression. Because these mutant APC lines failed to react not only with three different allele-specific but also monomorphic rat antimouse I-A mAbs, it is safe to conclude that they lost the expression of the I-A molecule itself, but not the expression of some of the I-A epitopes. They also lost the ability to stimulate KLH-reactive T cell clones of appropriate restriction specificity. Although we did not analyze these mutants further, these mutant APC lines would provide materials for the analysis of a variety of regulatory elements for class II expression. Preliminary analysis revealed that mutant 2C3 cells express Aad and Aj3z mRNA, but have a mutation in the Aad chain DNA sequence with alteration of amino acid residue at position 13 from valine to asparagine (our unpublished results). This might affect the association of Aad and A/3z for the

Table 2 Isolation of APC mutant lines Experiment

1 2 3 4 5 6 7 8 9 10 11 12 Total

Number of clones (% of total) Clones examined

APC function negative mutants

APC function negative class II negative mutants

APC function negative class II positive mutants

45 260 119 213 179 17.5 72 89 40 288 192 96 1768

15 176 26 41 106 106 72 11 12 139 0 59 763 (43.2%)

15 176 26 39 106 106 72 7 12 139 0 50 748 (42.3%)

0 0 0 2 0 0 0 4 0 0 0 9 15 (0.8%)

Mutant APC lines were isolated from parental MW4 B-B hybridoma AX lines or described in the materials and methods section. In 12 independent experiments, the mutant APC lines that lost the ability to stimulate auto-reactive A&z/Acrd-restricted lines that lost I-A expression and the number of mutant APC lines that express I-A but

TAPz class II-transfectant APC lines as number of cells examined, the number of T cell clones, the number of mutant APC lost T cell stimulation ability are indicated.

S. Koarada et al. /Journal

(A)

8F8

of Immunological Methods 182 (1995) 209-218

Ad. Mol.

Auto-Clone

KLH-Clone

215

h

4811 b

0

10

40

0

0

100

[aHI-TdR Uptake (Acpm X 109

(B)

100

MFl(%)

Ad. Mol.

Auto-Clone

KLH-Clone

TAfSz 2A10 8E6 2C8 2C3 13FlO 16A2 lB1

12E4 2All lG2 0

40

0

PHI-TdR Uotake

20

(Acpmx 109

0

100

0

100

MFN%)

Fig. 3. Characterization of mutant APC lines derived from (A) MW4 B-B hybridoma APC lines or (B) TApz transfectant AF’C lines. 1 x IO4 each parental or mutant APC lines were cultured with KLH-reactive A@/Aad-restricted T cell clone KGX44 (m), KGX58 ( q ) or auto-reactive Apz/Aad-restricted T cell clone KGUl08 (hatched), KGU140 (cross-hatched), KGUI 1.5t n ), KGX 109 ( q) for 24 h. Cytokines released into culture supematants were assayed by the proliferative responses of IL-2 and IL-4-dependent CT-6 cells. Wild type and mutant AF’C lines were stained with anti-I-A mAb BW/9 (hatched), anti-A/3z mAb 10.2.16 to), K24-199 CM), 4D5-12 (cross-hatched) and analyzed by FACScan. Mean fluorescence intensity (MFI) of each staining was expressed as a percentage of wild-type cells. Expression of adhesion molecules (Ad. Mol.) was measured by staining with anti-LFA-1 mAb FD441.8 ( l ) or anti-ICAM- mAb YN1/1.7 ( q ) and expressed as above.

216

S. Koarada et al. /Journal

of Immunological Methods 182 (1995) 209-218

expression of complete Acrd/APz heterodimer. That the importance of N-terminal amino acid sequences for the association of the class II dimer was reported by Sant et al. (1987). 3.5. Non-fimctional APC mutant cell lines with normal I-A molecule expression 4B8, 9A.5, 8F8 and 4Bll mutant APC lines derived from MW4 cells lost the ability to stimulate Apz/AcYd-specific autoreactive T cell clone KGU108 (Fig. 3A). These mutant APC lines also lost their stimulatory ability against other Apz/Aad-specific autoreactive T cell clones, as well as against Apz/Aaz-restricted KLH- and autoreactive T cell lines (data not shown). Also, a KLH-specific Apz/Aad-restricted T cell clone was not stimulated by these mutant APC lines. APC dose response and antigen dose response experiments demonstrated that these mutant APC lines do not stimulate T cell clones at any APC cell number or at any antigen concentrations (data not shown). In spite of this, they showed expression of normal amounts of I-A molecule, as examined by allele-specific and monomorphic anti-I-A mAbs. Preliminary DNA sequence analysis suggests that some of these mutant lines do not have any amino acid alterations of I-A molecules. The expression of ICAM- and LFA-1 is not deteriorated. The defect(s) in these mutant APC lines may reside in the pathway of peptide loading to class II molecules or in antigen processing.

3.6. Mutant APC lines that lost stimulatory activity against autoreactive but not against KLH-reactive T cell clones Mutant 2All and lG2 are of particular interest because these mutant lines could stimulate KLH-reactive, but not autoreactive T cell clones. Surface expression of class II, and ICY&I-l and LFA-1 is normal. The defect seems to reside somewhere in the unique processing/presentation pathways of self-antigens. DNA sequencing analysis revealed the nucleotide substitution with an altered deduced amino acid residue from alanine to threonine at position 69 of Aud chain of

the 2All class II molecule (Koarada et al., manuscript in preparation). This position was indicated to form a peptide binding pocket according to the crystallographic analysis of DRl molecule (Brown et al., 1993). This suggests that the binding of self-peptide recognized by autoreactive T cell clones is impaired by the alteration of amino acid residue at this position. 3.7. Miscellaneous mutant APC lines A mutant 16A2 APC line showed decreased expression of LFA-1 and almost no expression of ICAM-1. A mutant 1Bl APC line lost the ICAMmolecule completely but expressed normal amounts of LFA-1 molecule. In functional assays these mutant APC lines show the ability to stimulate autoreactive and KLH-reactive T cell clones, although the stimulatory activity is slightly low compared to the wild-type TApz. A mutant 18E5 line lost the expression of both ICAMand LFA-1. This mutant lost the ability to stimulate the KLH-reactive clone and some of the autoreactive clones almost completely. The autoreactive clone KGU140 was stimulated by this 18E5 mutant to some degree. This might be due to the relatively low dependency of the KGU140 clone to ICAM- molecule. In APC dose response experiments, these mutant APC lines showed stimulatory activity against all these T cell clones when high cell numbers of APCs were employed (data not shown). All these functional phenomena could reflect the loss or decreased expression of the ICAMmolecule, since it was reported that ICAM-l-defective APC mutant lines showed similar decreased ability to stimulate T cell hybridomas (Dang et al., 1990). A mutant APC clone 12E4 showed quite peculiar characteristics. This mutant was first isolated as a nonfunctional APC line with low I-A positive, but after subcloning turned out to be negative for the expression of I-A molecules. Curiously however, 12E4 mutant cells retained the ability to stimulate autoreactive T cell clone KGU140, although the stimulatory activity is somewhat weak. One possible explanation for this would be that this mutant APC line expresses limited species of self-antigens bound to APz/A&d molecules. The expression of

S. Koarada et al. /Journal

of Immunological Methods 182 (1995) X9-218

ICAM- and LFA-1 molecules on 12E4 was almost normal.

4. Discussion In this paper, we described the successful isolation of a variety of mutant APC lines by functional selection using cloned T cell lines. Experiments using hypothetical variant APC line TAPd (that expresses Apd/Acud) showed that the procedure described in this paper enabled us to enrich rare variant cells from less than 1% in the original APC mixtures. Repeated cycles of selection resulted in almost 100% of variant cell lines in the final APC populations. Based on these, we isolated mutant cell lines from EMS-treated TApz (that expresses Apz/Aad) or MW4 (that expresses Apz/Aad and A@/Aaz) B lymphoma APC lines. In a total of 12 independent isolations of mutant APC lines, more than 40% of APC clones lost the APC function to stimulate the autoreactive T cell clone KGU108. The efficiency of isolation of mutant cells varied from one experiment to another. In one experiment, all the APC clones lost the APC function, while in another experiment, no APC mutant lines were obtained (Table 2). However, the results, as a whole, may show high efficiency of this procedure for the isolation of mutant APC lines. The efficient isolation of mutant APC lines described in this paper depends on the specific growth inhibition of wild-type APC lines by cloned T cells. This efficient isolation was facilitated by the removal of large numbers of wild-type APC cells that aggregated with specific T cell clones. Experiments using hypothetical APC variant lines showed almost complete disappearance of wildtype APC lines after several cycles of selection by T cell clones. Although we were not able to demonstrate definitive killer activity of T cell clones against APC target cells by the conventional 4 h or 12 h Cr-release assay, the proliferation of APC lines was inhibited by specific recognition by T cell clones. It is suggested that one of the major pathways for class II-mediated CD4+ T cell cytotoxic activity is the interaction of Fas on target cells and FasL on T cells (Staider et al.,

217

1994). One reason for our failure to detect killer activity could be that the kinetics of killing is slow compared to conventional killer activity, probably due to the decreased Fas-mediated killing for this particular combination of T cell clones and APC lines. Alternatively, there exist other unidentified pathway(s) for class II-mediated kiher activity that might be employed in our present system. It should be possible to isolate mutant cell lines that have lost the expression of cell surface molecules involved in such unidentified killing pathway(s), if any, by the methods for mutant isolation described in this paper. One of the major differences for the isolation of APC mutant lines between our methods and those of Dang et al. (1990) is that we used cloned T cell lines instead of T cell hybridomas. Although some tedious steps are required for the maintenance of T cell clones our procedure has an advantage over that of Dang et al. because T cell clones will eventually be eliminated due to the absence of stimulation after the disappearance of wild-type APCs. The elimination of T cell clones from the culture will provide favorable conditions for the growth of mutant APCs and therefore will result in high efficiency isolation of mutant APC lines after limiting dilution cloning. Among these non-functional APC mutant lines, less than 1% expressed normal amounts of class II molecules as detected by mAbs. We characterized some such class II-positive non-functional APC mutant cell lines because these cell lines might have defect(s) somewhere in the entire pathway of antigen processing and presentation, as well as in the class II molecule structure. Initial characterization suggests that these include mutant APC lines with (11 altered MHC class II sequences, (2) loss of adhesion molecule expression, and (3) possible impairment of peptide loading. We show here an efficient method for isolating mutant APC lines and that initial characterizations of these mutant APC lines contained a variety of mutations in the entire pathways of antigen processing and presentation. Although the specificity of T cell clones analyzed in this paper is limited to self-antigens restricted by a limited set of class II specificities, this method

218

S. Koarada et al. /Journal of Immunological Methods 182 (1995) 209-218

should be applicable to any other antigen specificities and restriction specificities. These mutant APC lines with various characteristics should provide us with a variety of materials for the analysis of entire pathways of antigen processing and presentation. References Beck, B.N., Buerstedde, J.M., Krco, C.J., Nilson, A.E., Chase, C.G. and McKean, D.J. (1986) Characterization of cell lines expressing mutant I-Ab and I-Ak molecules allows the definition of distinct serologic epitopes on Aa and AP polypeptides. J. Immunol. 136, 2953. Brown, J.H., Jardetzky, T.S., Gorga, J.C., Stern, L.J., Urban, R.G., Strominger, J.L. and Wiley, D.C. (1993) Three-dimensional structure of the human class II histocompatibilty antigen HLA-DRl. Nature 364, 33. Cresswell, P. (1994) Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12, 259. Dang, L.H., Michalek, M.T., Takei, F., Benaceraf, B. and Rock, K.L. (1990) Role of ICAM- in antigen presentation demonstrated by ICAM- defective mutants. J. Immunol 144, 4082. Dang, L.H., Lien, L.L., Benacerraf, B. and Rock, K.L. (1993) A mutant antigen-presenting cell defective in antigen presentation expresses class II MHC molecules with an altered conformation. J. Immunol. 150, 4206. Germain, R.N. (1994) MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 76, 287. Germain, R.N. and Malissen, B. (1986) Analysis of the expression and function of class-11 major histocompatibility complex-encoded molecules by DNA-mediated gene transfer. Annu. Rev. Immunol. 4, 281. Glimcher, L.H. and Griffith, I.J. (1987) Mutations of class II MHC molecules. Immunol. today 8, 274. Glimcher, L.H., McKean, D.J., Choi, E. and Seidman, J.G. (1985) Complex regulation of class II gene expression: Analysis with class II mutant cell lines. J. Immunol. 135, 3542.

Gotoh, Y., Takashima, H., Noguchi, K., Nishimura, H., Tokushima, M., Shirai, T. and Kimoto, M. (1993) Mixed haplotype Apz/Aad class II molecule in (NZB x NZW)Fl mice detected by T cell clones. J. Immunol. 150, 4777. Kappler, J., Skidmore, B., White, J. and Marrack, P. (1981) Antigen-inducible, H-Zrestricted, interleukin 2-producing T cell hybridomas. Lack of independent antigen and H-2 recognition. J. Exp. Med. 153, 1198. Koch, N., Hiimmerling, G.J., Tada, N., Kimura, S. and Hammerling, U. (1982) Cross-blocking studies with monoclonal antibodies against I-A molecules of haplotypes b, d and k. Eur. J. Immunol. 12, 909. Lanier, L., Gutman, G., Lewis, D., Griswold, S. and Warner, N. (1982) Monoclonal antibodies against rat immunoglobulin kappa chains. Hybridoma 1, 125. Oi, V.T., Jones, P.P., Goding, J.W. and Herzenberg, L.A. (1978) Properties of monoclonal antibodies to mouse Ig allotypes, H-2, and Ia antigens. Curr. Top. Microbial. Immunol. 81, 115. Sant, A.J., Braunstein, N.S. and Germain, R.N. (1987) Predominant role of amino-terminal sequences in dictating efficiency of class II major histocompatibility complex aP dimer expression. Proc. Natl. Acad. Sci. USA 84, 8065. Sarmiento, M., Dialynas, D.P., Lancki, D.W., Wall, K.A., Lorber, MI., Loken, M.R. and Fitch, F.W. (1982) Cloned T lymphocytes and monoclonal antibodies as probes for cell surface molecules active in T cell-mediated cytolysis. Immunol. Rev. 68, 135. Stadler, B.M., Dougherty, S.F., Farrar, J.J. and Oppenheim, J.J. (1981) Relationship of cell cycle to recovery of IL 2 activity from human mononuclear cells, human and mouse T cell lines. J. Immunol. 127, 1936. Stalder, T., Hahn, S. and Erb, P. (1994) Fas antigen is the major target molecule for CD4+ T cell-mediated cytotoxicity. J. Immunol. 152, 1127. Takei, F. (1985) Inhibition of mixed lymphocyte response by a rat monoclonal antibody to a novel murine lymphocyte activation antigen (MALA-2). J. Immunol. 134, 1403. Tokushima, M., Koarada, S., Hirose, S., Gotoh, Y., Nishimura, H., Shirai, T., Miyake, K. and Kimoto, M. (1994) In vivo induction of IgG anti-DNA antibody by auto-reactive mixed haplotype A/3z/Aad MHC class II moleculespecific CD4+ T cell clones. Immunology 83, 221.