An efficient method for cloning human autoantigen-specific T cells

Journal of Immunological Methods 298 (2005) 83 – 92 www.elsevier.com/locate/jim

Research paper

An efficient method for cloning human autoantigen-specific T cells Stuart I. ManneringT, James A. Dromey, Jessica S. Morris1, Daniel J. Thearle, Kent P. Jensen, Leonard C. Harrison Autoimmunity and Transplantation Division, The Walter and Eliza Hall Institute of Medical Research, 1G Royal Parade, Parkville, Victoria 3050, Australia Received 14 October 2004; received in revised form 10 January 2005; accepted 14 January 2005 Available online 4 March 2005

Abstract T-cell clones are valuable tools for investigating T-cell specificity in infectious, autoimmune and malignant diseases. T cells specific for clinically-relevant autoantigens are difficult to clone using traditional methods. Here we describe an efficient method for cloning human autoantigen-specific CD4+ T cells pre-labelled with CFSE. Proliferating, antigen-responsive CD4+ cells were identified flow cytometrically by their reduction in CFSE staining and single cells were sorted into separate wells. The conditions (cytokines, mitogens and tissue culture plates) for raising T-cell clones were optimised. Media supplemented with IL-2+IL-4 supported growth of the largest number of antigen-specific clones. Three mitogens, PHA, anti-CD3 and antiCD3+anti-CD28, each stimulated the growth of similar numbers of antigen-specific clones. Cloning efficiency was similar in flat- and round-bottom plates. Based on these findings, IL-2+IL-4, anti-CD3 and round-bottom plates were used to clone FACSsorted autoantigen-specific CFSE-labelled CD4+ T cells. Sixty proinsulin- and 47 glutamic acid decarboxylase-specific clones were obtained from six and two donors, respectively. In conclusion, the CFSE-based method is ideal for cloning rare, autoantigen-specific, human CD4+ T cells. D 2005 Elsevier B.V. All rights reserved. Keywords: T-cell cloning; CFSE; Autoimmune disease; Type 1 diabetes; Glutamic acid decarboxylase; Proinsulin

Abbreviations: APC, Antigen-presenting cells; CFSE, 5,6-Carboxylfluorescein diacetate succinimidyl ester; DC, Dendritic cell; FACS, Fluorescence activated cell sorter; GAD, Glutamic acid decarboxylase; HLA, Human leukocyte antigen; HPLC, High performance liquid chromatography; IMDM, Iscove’s modified Dulbecco’s medium; PBMC, Peripheral blood mononuclear cells; PBS, Phosphate buffered saline; PHA, Phytohaemagluttinin; PI, Proinsulin; SI, Stimulation index; TT, Tetanus toxoid. T Corresponding author. Tel.: +61 3 9345 2457; fax: +61 3 9347 0852. E-mail address: [email protected] (S.I. Mannering). 1 Current address: Virology, Victorian Infectious Diseases Reference Laboratory, 10 Wreckyn St, North Melbourne, Victoria 3051, Australia. 0022-1759/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2005.01.001

84

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

1. Introduction Antigen-specific T-cell clones are valuable tools for mapping T-cell epitopes. However, human autoantigen-specific T-cell clones have been particularly difficult to isolate due to their rarity in blood. Alloantigen-specific T-cell clones were first isolated by antigen stimulation followed by limiting dilution or soft agar cloning (Bach et al., 1979). The limiting dilution method has been used for many years to clone alloantigen- and microbial vaccine-specific T cells present at high frequency in blood (Pawelec and Wernet, 1980). This method has been improved, but only marginally. A combination of irradiated peripheral blood mononuclear cells (PBMC) and the EBVtransformed B-cell line, JY, supported the growth of T-cell clones better than JY-cells alone (Spits et al., 1982b), and stimulation with PHA further increased the cloning efficiency (Malissen et al., 1981; Spits et al., 1982a). Activation of T cells with autologous DC has also been reported to improve cloning efficiency (Fonteneau et al., 2001). Despite these improvements, T-cell cloning by limiting dilution remains a difficult and time-consuming task. More recently, the advent of soluble HLA tetramers has facilitated T-cell cloning by allowing direct identification of antigen-specific T cells at the outset. For example, human CD8+ T cells specific for melanoma antigens (Dunbar et al., 1998, 1999) and CD4+ T cells specific for the autoantigen glutamic acid decarboxylase-65 (GAD) (Novak et al., 2001b) have been cloned by sorting tetramer-positive cells. However, tetramer construction is time-consuming and requires a detailed knowledge of both the HLA molecule and of the peptide epitope to which it binds. In the absence of this information, tetramers could be made using a panel of peptides spanning the entire antigen of interest. This approach was used to isolate Herpes Simplex Virus-2, VP16-specific CD4+ T-cell clones with HLA DRB1*0401 tetramers complexed to a panel of VP16 peptides (Kwok et al., 2001; Novak et al., 2001a). Combining HLA tetramer and peptide libraries partially overcomes the problem of epitope identification. However, T cells that recognise peptides bound weakly to HLA may be overlooked, synthesis of peptide libraries is expensive, peptide purity can vary dramatically and recombinant HLA molecules still need to be produced for each HLA allele.

Type 1 diabetes is an autoimmune disease in which proinsulin appears to play a major role as an autoantigen in driving T-cell mediated destruction of pancreatic h cells (Narendran et al., 2003). There have been only two reports of (pro)insulin-specific human CD4+ T-cell clones (Schloot et al., 1998; Semana et al., 1999) in type 1 diabetes, each describing preliminary analysis of just a single clone. Hence, our aim was to develop a better method for cloning T cells specific for autoantigens. Here we describe a simple, robust and generic method for cloning T cells applied to T cells specific for proinsulin and the other h-cell antigen, glutamic acid decarboxylase.

2. Methods 2.1. Blood donors and PBMC isolation Blood was obtained by venepuncture from volunteer donors with informed consent and ethics committee approval. PBMC were isolated over Ficoll/ Hypaque (Amersham Pharmacia Biotech AB, Uppsala Sweden) and washed twice in phosphate buffered saline (PBS). Cells were cultured in Iscove’s modified Dulbecco’s medium (IMDM) (Gibco, Rockville, MA, USA) supplemented with 5% pooled male human serum, 2 mM glutamine (Glutamax)(Gibco), 510 5 M 2-mercaptoethanol (Sigma, St Louis, MO, USA), penicillin (100 U/ml), streptomycin (100 Ag/ ml) and 100 AM non-essential amino acids (Gibco), referred to as complete culture medium. 2.2. Antigens Tetanus toxoid (TT) was supplied by CSL (Parkville, Victoria, Australia). Recombinant GAD was produced in Baculovirus and purified as described (Bach et al., 1979). Endotoxin concentration of the GAD stock solution, measured by Limulus lysate assay (BioWhittaker, Walkerville, MD,USA), was 1.2 EU/mg/ml. Recombinant human proinsulin was produced in-house as published (Cowley and Mackin, 1997). Briefly, after refolding and reversed phase high performance liquid chromatography (RP-HPLC) purification, the protein resolved as a single species of correct molecular weight in matrix-assisted laser desorption/ionisation-time of flight mass (MALDI-

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

TOF) spectrometry. The endotoxin concentration of proinsulin stock solution was 0.51 EU/mg/ml. 2.3. CFSE staining, cell culture and sorting PBMC (1107/ml in PBS) were incubated at 37 8C for 5 min with 0.5 AM CFSE (stock at 0.5 mM in DMSO) (Molecular Probes, Eugene, OR, USA). Staining was terminated by adding culture medium containing 5% pooled human serum, the cells were washed once in PBS/1% pooled human serum and resuspended in culture medium at 1.33106/ml. Stained cells (2105/well, 150 Al) were cultured in 96-well round-bottom plates (Becton Dickinson Labware, Franklin Lakes, NJ, USA) with medium alone or with TT (final concentration, 10 LfU/ml), GAD (5 Ag/ml) or proinsulin (10 Ag/ml). Unstained cells included in all experiments were used to set the compensations on the flow cytometer. After 7 days of culture, cells for each antigen were pooled, washed in PBS and stained on ice with anti-human CD4-PE (IgG2a, clone RPA-T4), (BD Pharmingen, San Diego, CA, USA). Optimal compensation and gain settings were determined for each experiment based on unstained and single-stained samples. A single, viable (propidium-iodide negative), CD4+, CFSEdim cell was sorted into each well of a 96-well plate, except for the outside edge wells, which contained 200 Al of water. Each well contained feeder cells, cytokines and mitogen in 100 Al of culture medium as described below. Irradiated (2000 rad) fresh or thawed PBMC (1105) from two unrelated donors and 5104 irradiated (5000 rad) EBV-transformed JY cells were used as feeder cells in the cloning plates. In some experiments irradiated (1000 rad) Jurkat cells (2105 cells/well) were used instead of PBMC. The following cytokines (Peprotech, Rocky Hill, NJ, USA) were used at the final concentrations indicated: IL-2 (10 U/ml), IL-4 (5 ng/ml) IL-7 (5 ng/ml) and IL-15 (5 ng/ml). The following mitogens were used at the final concentrations indicated: PHA (2.5 Ag/ml), anti-CD3 (clone OKT-3, 30 ng/ml) and anti-CD28 (clone L293, low azide, low endotoxin; 100 ng/ml, BD-Pharmingen, San Jose, CA, USA). Fungizone (Amphotericin B, Bristol-Myers Squibb, Princeton, NJ, USA) was added to all cultures at a final concentration of 2 Ag/ml. Cells were fed every 7 days with fresh cytokines in 50 Al of medium, to the final concentrations indicated. Grow-

85

ing clones were identified by visual inspection, expanded into 48-well plates after ~2 weeks and then tested for antigen specificity by 3H-thymidine incorporation assay. Antigen-specific clones (SIN3, see below) were expanded with PHA, IL-2+IL-4 and feeder cells as described above, or with anti-CD3 as described (Riddell and Greenberg, 1990), except IL2+IL-4 were used instead of IL-2 alone. 2.4. 3H-thymidine incorporation assay Clones were tested for antigen specificity using the H-thymidine incorporation assay. APC were autologous PBMC that had been stored in liquid nitrogen, thawed and irradiated (2000 rad). Each clone was tested in duplicate against autologous APC (~1105 cells/well) with and without antigen. Cultures were set up in 96-well round-bottom plates, in complete culture media. T-cell clones were resuspended and half the volume (usually 200 Al) was transferred to a 10 ml tube. The T cells were washed, first in PBS and then in 0.5 ml of complete culture medium and finally resuspended in 450 Al of complete culture medium. Cells were not counted, but 100 Al of washed, cloned, T cells were added to wells containing APC with and without antigen. After 2 days, 3H-thymidine (0.5 ACi/ well) was added for 18 h after which the cells were harvested and incorporated radioactivity measured by h-scintillation counting. Clones that had a stimulation index (SI, cpm with antigen/cpm without antigen) of N3 were considered to be antigen-specific. Results were expressed as the cloning efficiency, i.e. the number of antigen-specific clones as a percentage of the number of wells plated. 3

2.5. PCR for TCR Vb genes TCR Vh usage of the clones was determined by PCR. Briefly, RNA was isolated from 0.5–1106 cloned T cells and 0.5 Ag of RNA was reverse transcribed to cDNA with Moloney Murine Leukemia Virus Reverse Transcriptase (M-MLV RT Promega, WI, USA) in a final volume of 40 Al. The RT reaction was diluted 1:1.25 and 1 Al used as template for the PCR (94 8C for 5 min, then 30 cycles each at 94 8C, 10 s; 55 8C 60 s, 72 8C 60 s, then finally 72 8C for 5 min) using the optimised Vh primers (10 AM) and PCR conditions described (Genevee et al., 1992).

86

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

3. Results

3.2. Optimisation of cytokines

3.1. Identification of antigen-specific CD4+T cells

To determine the optimal combination of cytokines for growth of T-cell clones the efficiency of TT-specific T-cell cloning in IL-2, IL-2+IL-4, IL2+IL-7, IL-2+IL-15, IL-2+IL-4, -7, -15 was compared. A summary of the results from three separate experiments is shown (Table 1). The combination of IL-2+IL-4 supported growth of the most clones, with an average cloning efficiency of 13.6%. The addition of IL-7+IL-15 did not increase cloning efficiency (10.7%). No cells grew in control wells that contained only irradiated feeder cells. Based on these data, IL-2+IL-4 were used for subsequent cloning experiments.

Antigen-specific T cells were identified by the reduction in their CFSE staining during culture with antigen. Gates on the flow cytometer were set to exclude dead cells and doublets (Fig. 1). All experiments included a no antigen control. One CD4+, CFSEdim cell was sorted into each of the 60 inner wells of a 96-well plate that contained cytokines, mitogen and feeder cells. TT was used as a model antigen to optimise cytokines, mitogen and well shape for cloning human antigen-specific CD4+ T cells.

Fig. 1. Flow cytometric identification of antigen-specific CD4+ T cells. Forward (FSC) and side scatter (SSC) gates were set to include mononuclear cells (A) and exclude propidium-iodide (PI) stained dead cells (B). Doublets were excluded based on FSC width (FSC-W) (C). Cells that fell in these gates were examined for CFSE and CD4-PE staining. In each experiment the level of proliferation in the absence of antigen was determined (D). Representative plots, including the percentage of CD4+ CFSEdim cells, for tetanus toxoid (E) and proinsulin (F) cultures are shown. A single CD4+ CFSEdim cell was sorted into each of the inner 60 wells of a 96-well plate.

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

87

Table 1 Summary of cytokines tested Cytokines

Expt

Number of wells plated

Number of growing clonesa

Number of antigen specific clonesb

Cloning efficiency (%)c

Average cloning efficiency (%)

IL-2

1 2 3

NDd 60 120

ND 2 8

ND ND 8

ND ND 6.6

b5.0e

IL-2, -4

1 2 3

57 60 60

11 15 14

9 1 14

15.8 1.7 23.3

13.6

IL-2, -7

1 2 3

60 60 60

0 5 3

0 ND 2

0 ND 3.3

b3.9

IL-2, -15

1 2 3

60 60 60

1 2 1

ND ND ND

ND ND ND

b2.2

IL-2, -4, -7, -15

1 2 3

42 60 60

7 15 14

3 1 14

7.1 1.7 23.3

10.7

a b c d e

Number of wells with clones growing after 2 weeks of culture. Number of antigen-specific clones by 3H-thymidine incorporation. Number of antigen-specific clones as a percentage of the number of wells plated. ND, not determined. Where clones were not tested for specificity, efficiency was calculated assuming that all growing clones were antigen-specific.

3.3. Optimisation of mitogen To optimise the mitogen, the cloning efficiencies with PHA, anti-CD3 mAb, and anti-CD3+anti-CD28 mAbs were compared (Table 2). The average

cloning efficiencies with PHA, anti-CD3 and antiCD3+anti-CD28 were 8.6%, 11.6% and 11.9%, respectively. As addition of anti-CD28 made no difference, anti-CD3 alone was used for subsequent experiments.

Table 2 Summary of mitogen comparison Mitogen

Expt

Number of wells plated

Number of growing clones

Number of antigen-specific clones

Cloning efficiency (%)

Average cloning efficiency (%)

PHA

1 2 3 4

60 114 120 120

16 3 12 21

9 2 11 10

15.0 1.8 9.2 8.3

8.6

CD3

1 2 3 4

120 114 120 120

26 12 21 26

15 9 14 17

12.5 7.9 11.7 14.2

11.6

CD3/CD28

1 2 3 4

ND 114 120 120

ND 21 12 24

ND 13 10 19

ND 11.4 8.3 15.8

11.9

88

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

3.4. Optimisation of well shape and feeder cells The optimal well shape was determined by comparing the cloning efficiency in round- and flatbottom tissue culture plates (Table 3). Cloning efficiency was similar in flat- (13.7%), and roundbottom (15.4%) plates. Hence, round-bottom plates continued to be used. We compared the efficiency of TT-specific T-cell clone isolation when irradiated PBMC or the T-cell line Jurkat were used as feeder cells (Table 4). PBMC-feeder cells lead to the growth of more antigen-specific clones (15.4%) than Jurkat (12.9%). Thus clones could be isolated using Jurkat as feeder cells, but PBMC resulted in more efficient cloning and continued to be used. Furthermore, in preliminary experiments, thawed PBMC that had been stored in liquid nitrogen, then irradiated, could be substituted for fresh, irradiated PBMC as feeder cells, with no change in cloning efficiency (data not shown). 3.5. Isolation of autoantigen-specific CD4+ T-cell clones Using the conditions determined above a total of 60 proinsulin-specific clones were isolated from two healthy donors, two diabetic donors and two donors with islet autoantibodies (Table 5). Forty-seven GADspecific CD4+ T-cell clones were isolated from two healthy donors (Table 5). Clones were considered to be antigen specific if the SI was N3.0. Representative results, from antigen-specificity testing, are shown in Table 6. PCR amplification of TCR Vh genes was used to obtain confirmation that the clones expressed a single TCR Vh gene. Each PCR reaction yielded a band derived from the Ch region. Where a Vh TCR rearrangement was present a second product was amplified (Fig. 2). Twelve clones from three inde-

pendent cloning experiments were tested, each had a single rearranged TCR Vh and derived from a single cell. Most (~80%) of the clones could be grown to large numbers (107–108) either with repeated stimulation by PHA, IL-2+IL-4 and feeder cells, or by anti-CD3 mediated growth (Riddell and Greenberg, 1990). The analysis of these clones will be reported elsewhere.

4. Discussion Here we describe an efficient method for cloning human CD4+ T cells. Rare, antigen-specific T cells were identified by their proliferation and resultant decrease in CFSE staining in response to antigen, and cloned after sorting 1 cell/well using a FACS sorter. A previous report described the optimisation of a CFSEbased proliferation assay and the phenotype of the divided cells (Mannering et al., 2003a). Here we have optimised the conditions for growing antigen-specific clones after sorting antigen-responsive T cells. Cloning human autoantigen-specific T cells has been a long-standing technical challenge. The major problem is that the frequency of responding T cells in blood is low and repeated stimulation may lead to over-growth of T cells specific for irrelevant antigens. The CFSE-based method allows rare antigen-responsive cells to be identified based on their proliferative response to antigen, and cloned directly. Our method has several advantages. First, clones can be generated rapidly. Because repeated re-stimulation with antigen and APC is not required clones can be generated within 3 weeks and expanded to large numbers (107–108) after a further 2 weeks. Second, unlike tetramers, any form of antigen (protein, peptide, intact cells, cell lysates) can be used. In most of our experiments, intact, recombinant protein was used, but

Table 3 Comparison of plates Well shape

Expt

Number of wells plated

Number of growing clones

Number of antigen-specific clones

Cloning efficiency (%)

Average cloning efficiency (%)

Flat

1 2

120 120

14 22

14 19

11.6 15.8

13.7

Round

1 2

120 120

18 20

18 19

15.0 15.8

15.4

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

89

Table 4 Comparison of feeder cells Feeder cells

Expt

Number of wells plated

Number of growing clones

Number of antigen-specific clones

Cloning efficiency (%)

Average cloning efficiency (%)

Jurkat

1 2

120 120

5 26

5 26

4.2 21.6

12.9

PBMC

1 2

120 120

18 20

18 19

15.0 15.8

15.4

we have also used synthetic peptide for the initial stimulation and isolated peptide-specific clones (data not shown). Third, tetramers are not required. Fourth, the antigen-responsive population may be further characterised based on surface marker or cytokine expression and the sub-population of interest cloned. Methods for staining viable, cytokine-secreting cells have been developed (Manz et al., 1995; Brosterhus et al., 1999) which are suitable for this application. Fifth, lineages other than CD4+ T cells can be cloned with the CFSE method. We have cloned CD8+ T cells specific for the HLA A2-restricted, matrix protein 58–66 epitope (data not shown). Sixth, this method is particularly advantageous for cloning rare T cells. Few studies have attempted to determine the optimum conditions for growing T-cell clones. We compared T-cell clone growth in the presence of the gamma common chain of cytokines, IL-2, -4, -7 and -15, that regulate T-cell growth and survival (Li et al., 2001; Schluns and Lefrancois, 2003). IL-2 has been widely used to support T-cell growth in cloning experiments. However, in our experiments IL-2 alone was inferior to IL-2+IL-4. Nonetheless, clones could be isolated in IL-2 alone if required. It is possible, but we believe unlikely, that higher doses of IL-2

might better support T-cell growth during the time required for clones to reach sufficient numbers for antigen-specificity testing. We avoided high (z50 U/ ml) concentrations of IL-2 because in our experience these lead to a brief period of rapid growth followed by cell death. Of all the parameters tested, cytokines had the most effect on cloning efficiency. However, it should be noted that exposure to different combination of cytokines may alter the phenotype of the clones isolated. If the surface phenotype or function of the clones is critical to the end-use, clones of the desired function or phenotype can be selected after initial antigen-specificity testing. Previously, we described the optimal conditions for detecting CD4+ T-cell responses to autoantigens using CFSE (Mannering et al., 2003a). We found that 5 AM CFSE inhibited the CD4+ T-cell response to TT, whereas 0.5 AM allowed divided cells to be detected readily. Because batches of CFSE could vary we would recommend that the optimal concentration for each batch is determined by titration in a TT-driven proliferation assay. The yield of antigen-specific clones differed between experiments. This is not surprising, given that in outbred human populations there will be

Table 5 Summary of islet autoantigen-specific CD4+ T-cell clones Antigen

Expt

Clinical status of donor

Number of wells plated

Number of antigen-specific clones

Cloning efficiency (%)

Average cloning efficiency (%)

Proinsulin

1 2 3 4 5 6

Diabetic Pre-diabetic Healthy Healthy Diabetic Pre-diabetic

480 540 480 480 480 360

17 11 3 1 3 25

3.5 2.0 0.6 0.2 0.6 6.9

2.3

GAD

1 2

Healthy Healthy

480 480

43 4

9.0 0.8

4.9

90

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

Table 6 Sample raw data from specificity testing assays Antigen

Clone number

No antigen replicate 1

No antigen replicate 2

Antigen replicate 1

Antigen replicate 2

SIa

TTb

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

264c 450 362 326 572 648 334 750 327 33,030 406 16,489

331 343 179 158 149 169 305 612 564 30,586 520 13,912

11,202 9939 1658 769 2103 3128 6387 678 8963 23,514 7991 13,912

10,132 9891 2090 1029 2030 2615 8221 639 9168 27,434 7196 14,167

35.9 25.0 6.9 3.7 5.7 7.0 22.9 1.0 20.3 0.8 16.4 0.9

Proinsulind

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

97 108 274 15,149 196 216 71 99 313 172 128 85

129 173 335 17,889 203 220 100 166 356 115 141 69

151 1638 2289 15,910 1441 3983 92 293 3643 174 153 173

163 912 1624 13,293 1781 6379 62 306 2293 103 131 176

1.4 9.1 6.4 0.9 8.1 23.8 0.9 2.3 8.9 1.0 1.1 2.3

GADe

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

458 484 350 373 217 282 162 207 2963 305 125 547

218 196 231 225 271 215 191 144 1517 147 150 397

211 166 3819 7978 220 267 3092 144 1683 247 386 175

243 137 3816 6647 173 358 3450 179 3401 228 248 195

0.7 0.4 13.1 24.5 0.8 1.3 18.5 0.9 1.1 1.1 2.3 0.4

a Stimulation index (SI) is mean cpm with antigen divided by the mean cpm without antigen. SIN3.0 are considered to be positive and are shown in bold. b First 12 clones tested from Table 1, expt 3. c Cpm from 3H-thymdine incorporation assay. d First 12 clones tested from Table 5, proinsulin expt 6.

considerable genetic heterogeneity between donors. The ratio of the number of cells that divided in response to antigen compared to the number that divided in the absence of antigen (known as the cell division index or CDI) impacts directly upon the cloning efficiency. Indeed, it may not be possible to clone T cells in experiments where strong antigenindependent proliferation, or very weak responses to

antigen are observed. Variability in CDI between experiments will result in variability in the yield of clones between experiments. Furthermore, T cells could differ in their responses to cytokines, mitogens and other components of culture media. Such variability will affect the efficiency of cloning. Since this variability is unavoidable in human immunological research, we optimised conditions for T-cell cloning

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

91

Fig. 2. Analysis of clonality by TCR Vh expression. The TCR Vh expression of cloned T cells was determined by PCR using a panel of optimised primers specific for 24 members of the Vh family. Each reaction contains primers specific for the constant region (Ch) and a single Vh specific primer. The Ch band serves as a positive control. A single Vh product was detected in each of 12 clones tested. In the example shown, the clone expresses Vh6. Lane dMT contains molecular weight markers and lane dWT contains water instead of template cDNA.

by testing several donors. Using these conditions we successfully isolated autoantigen-specific clones from eight different donors. Recombinant protein antigens were used to stimulate T cells. Unfortunately, recombinant proteins, like peptides (Mannering et al., 2003b), are rarely completely free from modified forms or contaminants. For example, our initial attempts to clone proinsulinspecific T cells lead to the isolation of clones specific for a minor contaminant in the synthetic peptide (Mannering et al., 2003b). Our proteins did not contain significant levels of LPS, or other contaminants detectable by mass spectroscopy. Nonetheless, to rigorously prove the specificity of T-cell responses the same cells should be shown to respond to another form of the same antigen, such as a peptide. We have confirmed the specificity of some of the clones described here and this work is continuing. CFSE-based T-cell cloning relies upon detecting antigen-responsive cells that have divided in response to cognate antigen. Consequently, the magnitude of the T-cell response will affect the efficiency of the cloning. Previously we found responses to proinsulin were weaker than to GAD (Mannering et al., 2003a). Here we found the yield of proinsulin-specific clones

was lower than to GAD. Addition of anti-CD28 has been reported to enhance the response to GAD and proinsulin peptides in ELISpot assays (Ott et al., 2004a,b). We have not attempted to augment proinsulin-specific proliferation with exogenous cytokines or costimulatory signals. Our aim was to clone T cells that proliferate in response to autoantigens under conditions that were as close to physiological as possible. In some situations, addition of reagents to augment T-cell proliferation may be required to elicit sufficient antigen-responsive cells for cloning. We have previously shown that CD8+, CD19+ and CD4+ cells proliferate in the same cultures (Mannering et al., 2003a). Hence, our method could also be applied to cloning these cells. We anticipate that with the approach described here, cloning of rare autoantigen-specific T cells will become routine.

Acknowledgments This work was funded by a Center Program Grant from the Juvenile Diabetes Research Foundation (JDRF) and by the National Health and Medical Research Council (NHMRC) of Australia.

92

S.I. Mannering et al. / Journal of Immunological Methods 298 (2005) 83–92

JPK was supported by a JDRF Post-doctoral Fellowship (3-2002-638).

References Bach, F.H., Inouye, H., Hank, J.A., Alter, B.J., 1979. Human T lymphocytes clones reactive in primed lymphocyte typing and cytotoxicity. Nature 281, 307 – 309. Brosterhus, H., Brings, S., Leyendeckers, H., Manz, R.A., Miltenyi, S., Radbruch, A., Assenmacher, M., Schmitz, J., 1999. Enrichment and detection of live antigen-specific CD4(+) and CD8(+) T cells based on cytokine secretion. Eur. J. Immunol. 29, 4053 – 4059. Cowley, D.J., Mackin, R.B., 1997. Expression, purification and characterization of recombinant human proinsulin. FEBS Lett. 402, 124 – 130. Dunbar, P.R., Ogg, G.S., Chen, J., Rust, N., van der Bruggen, P., Cerundolo, V., 1998. Direct isolation, phenotyping and cloning of low-frequency antigen-specific cytotoxic T lymphocytes from peripheral blood. Curr. Biol. 8, 413 – 416. Dunbar, P.R., Chen, J.L., Chao, D., Rust, N., Teisserenc, H., Ogg, G.S., Romero, P., Weynants, P., Cerundolo, V., 1999. Cutting edge: rapid cloning of tumor-specific CTL suitable for adoptive immunotherapy of melanoma. J. Immunol. 162, 6959 – 6962. Fonteneau, J.F., Larsson, M., Somersan, S., Sanders, C., Munz, C., Kwok, W.W., Bhardwaj, N., Jotereau, F., 2001. Generation of high quantities of viral and tumor-specific human CD4+ and CD8+ T-cell clones using peptide pulsed mature dendritic cells. J. Immunol. Methods 258, 111 – 126. Genevee, C., Diu, A., Nierat, J., Caignard, A., Dietrich, P.Y., Ferradini, L., Roman-Roman, S., Triebel, F., Hercend, T., 1992. An experimentally validated panel of subfamily-specific oligonucleotide primers (V alpha 1-w29/V beta 1-w24) for the study of human T cell receptor variable V gene segment usage by polymerase chain reaction. Eur. J. Immunol. 22, 1261 – 1269. Kwok, W.W., Gebe, J.A., Liu, A., Agar, S., Ptacek, N., Hammer, J., Koelle, D.M., Nepom, G.T., 2001. Rapid epitope identification from complex class-II-restricted T-cell antigens. Trends Immunol. 22, 583 – 588. Li, X.C., Demirci, G., Ferrari-Lacraz, S., Groves, C., Coyle, A., Malek, T.R., Strom, T.B., 2001. IL-15 and IL-2: a matter of life and death for T cells in vivo. Nat. Med. 7, 114 – 118. Malissen, B., Charmot, D., Mawas, C., 1981. Expansion of human lymphocyte populations expressing specific immune reactivities: III. Specific colonies, either cytotoxic or proliferative, obtained from a population of responder cells primed in vitro. Preliminary immunogenetic analysis. Hum. Immunol. 2, 1 – 13. Mannering, S.I., Morris, J.S., Jensen, K.P., Purcell, A.W., Honeyman, M.C., van Endert, P.M., Harrison, L.C., 2003a. A sensitive method for detecting proliferation of rare autoantigen-specific human T cells. J. Immunol. Methods 283, 173 – 183.

Mannering, S.I., Purcell, A.W., Honeyman, M.C., McCluskey, J., Harrison, L.C., 2003b. Human T-cells recognise N-terminally Fmoc-modified peptide. Vaccine 21, 3638 – 3646. Manz, R., Assenmacher, M., Pfluger, E., Miltenyi, S., Radbruch, A., 1995. Analysis and sorting of live cells according to secreted molecules, relocated to a cell-surface affinity matrix. Proc. Natl. Acad. Sci. U. S. A. 92, 1921 – 1925. Narendran, P., Mannering, S.I., Harrison, L.C., 2003. Proinsulin—a pathogenic autoantigen in type 1 diabetes. Autoimmun. Rev. 2, 204 – 210. Novak, E.J., Liu, A.W., Gebe, J.A., Falk, B.A., Nepom, G.T., Koelle, D.M., Kwok, W.W., 2001a. Tetramer-guided epitope mapping: rapid identification and characterization of immunodominant CD4+ T cell epitopes from complex antigens. J. Immunol. 166, 6665 – 6670. Novak, E.J., Masewicz, S.A., Liu, A.W., Lernmark, A., Kwok, W.W., Nepom, G.T., 2001b. Activated human epitope-specific T cells identified by class II tetramers reside within a CD4high, proliferating subset. Int. Immunol. 13, 799 – 806. Ott, P.A., Berner, B.R., Herzog, B.A., Guerkov, R., Yonkers, N.L., Durinovic-Bello, I., Tary-Lehmann, M., Lehmann, P.V., Anthony, D.D., 2004a. CD28 costimulation enhances the sensitivity of the ELISPOT assay for detection of antigenspecific memory effector CD4 and CD8 cell populations in human diseases. J. Immunol. Methods 285, 223 – 235. Ott, P.A., Dittrich, M.T., Herzog, B.A., Guerkov, R., Gottlieb, P.A., Putnam, A.L., Durinovic-Bello, I., Boehm, B.O., Tary-Lehmann, M., Lehmann, P.V., 2004b. T cells recognize multiple gad65 and proinsulin epitopes in human type 1 diabetes, suggesting determinant spreading. J. Clin. Immunol. 24, 327 – 339. Pawelec, G., Wernet, P., 1980. Restimulation properties of human alloreactive cloned T-cell lines. Dissection of HLA-D-region alleles in population studies and in family segregation analysis. Immunogenetics 11, 507 – 519. Riddell, S.R., Greenberg, P.D., 1990. The use of anti-CD3 and antiCD28 monoclonal antibodies to clone and expand human antigen-specific T cells. J. Immunol. Methods 128, 189 – 201. Schloot, N.C., Willemen, S., Duinkerken, G., de Vries, R.R., Roep, B.O., 1998. Cloned T cells from a recent onset IDDM patient reactive with insulin B-chain. J. Autoimmun. 11, 169 – 175. Schluns, K.S., Lefrancois, L., 2003. Cytokine control of memory Tcell development and survival. Nat. Rev., Immunol. 3, 269 – 279. Semana, G., Gausling, R., Jackson, R.A., Hafler, D.A., 1999. T cell autoreactivity to proinsulin epitopes in diabetic patients and healthy subjects. J. Autoimmun. 12, 259 – 267. Spits, H., Borst, J., Terhorst, C., de Vries, J.E., 1982a. The role of T cell differentiation markers in antigen-specific and lectindependent cellular cytotoxicity mediated by T8+ and T4+ human cytotoxic T cell clones directed at class I and class II MHC antigens. J. Immunol. 129, 1563 – 1569. Spits, H., Ijssel, H., Terhorst, C., de Vries, J.E., 1982b. Establishment of human T lymphocyte clones highly cytotoxic for an EBV-transformed B cell line in serum-free medium: isolation of clones that differ in phenotype and specificity. J. Immunol. 128, 95 – 99.