TAP1^–/– mice present oligoclonal BV-BJ expansions following the rejection of grafts bearing self antigens

Abstract

Our previous work showed that transporter associated with antigen processing 1 (TAP1)^–/– (H-2^b) mice rejected grafts from H-2^b mice which display a normal density of class I major histocompatibility complex (MHC) molecules at the cell surface. Our results indicated that H-2^b molecules themselves may be a target in this kind of rejection and that CD4⁺ T cells play a major role in this autoreactive process. Our data also suggested that TAP1^–/– mice, in addition to the well-recognized phenotype of class I and CD8⁺ T-cell deficiency, present a functional alteration in their autoreactive CD4⁺ T-cell repertoires. In this model of inflammatory autoreactivity to modified self, we have analysed T-cell receptor (TCR) V-beta–J-beta (BV-BJ) usage by complementarity determining region 3 (CDR3) length spectratyping in splenocytes from naïve TAP1^–/– mice and transplanted TAP1^–/– mice that rejected B6 heart grafts or responded to synthetic self H-2K^b peptides. Importantly, oligoclonal T-cell expansions shared by different animals were detected in the peripheral T-cell repertoire of transplanted TAP1^–/– mice. Such public expansions were also induced in vitro by H-2K^b peptides, suggesting that dominant class I peptides can induce preferential expansions of restricted T-cell populations during rejection. Some of these public T-cell expansions were also detected in transplanted mice even before in vitro stimulation with peptides, indicating that post-transplantation expansion of these populations had occurred in vivo. The functional activity of these T-cell populations awaits elucidation, as do the underlying mechanisms involved in the inflammatory autoreactive process, in TAP1^–/– mice.

Introduction

The transporter associated with antigen processing (TAP) 1 and TAP2 molecules are responsible for the transport of peptides from the cytosol into the endoplasmic reticulum, where they bind to major histocompatibility complex class I molecules.^1,2 As a consequence of targeted deletion of the TAP1 gene, TAP1 mutant mice (TAP1^–/–) have a deficiency in peptide transport across the endoplasmic reticulum and a greatly reduced expression of MHC class I antigens on the cell surface.^3,4 Thymic positive selection of CD8⁺ T cells is therefore altered in TAP1^–/– mice, which mature with very few CD8⁺ T cells in the periphery but have normal numbers of CD4⁺ T cells and natural killer (NK) cells.^3,5 These few positively selected CD8⁺ T cells have been reported to recognize alloantigens and tumour cells.^4,6–8 It has also been shown that TAP1^–/– mice have extrathymic CD8⁺ T cells in the liver. However, these extrathymic CD8⁺ T cells do not seem to migrate to the periphery and have a limited ability to respond to allogeneic stimulation.⁹

Our previous study showed that H-2^b+ class I-deficient TAP1^–/– mice reject grafts from H-2^b+ donors [TAP^+/+, C57BL/6 (B6), 129/Sv, and TAP1⁺ littermates] that express a normal density of MHC class I molecules at the cell surface.¹⁰ We also showed that H-2^b class I molecule or derived peptides are targets in the rejection of B6 grafts by TAP1^–/–, as presensitization of TAP1^–/– mice by subcutaneous immunization with synthetic peptides derived from the sequence of the H-2K^b class I molecule with complete Freud's adjuvant (CFA), but not with adjuvant alone or with an irrelevant peptide, significantly accelerated the rejection of B6 skin grafts.¹¹

Although the dominant view in the literature is that the rejection of H-2^b+ grafts by MHC class I-deficient mice is mediated by CD8⁺ T cells and that it should be considered as a minor antigen disparity model,^4,7,12 our previous studies support an important role for CD4⁺ T cells in this model of graft rejection. We showed that in vivo depletion of CD4 T cells in TAP1^–/– mice induced a significant prolongation of B6 skin graft survival¹¹ and that graft-infiltrating lymphocytes in B6 grafts rejected by TAP1^–/– mice were mostly CD4⁺ T cells. We interpreted the B6 graft rejection in this model as an autoimmune response in view of the fact that the donors and recipients display neither major nor minor histocompatibility disparities and that H-2^b MHC class I-derived peptides are involved in the trigger of B6 graft rejection by TAP1^–/– mice.

We have been working on the hypothesis that TAP1^–/– mice may have a functionally altered CD4 T-cell repertoire, in addition to the reported CD8 T-cell deficiency. The molecular basis for the putative dysfunction of the T-cell repertoire towards inflammatory autoimmunity is yet to be clarified. We hypothesized that these T cells have a skewed repertoire presumably resulting from deficiency of Kb expression during selection in the thymus. Consistent with this idea, effector CD4⁺ T cells from TAP1^–/– mice could be activated to develop an inflammatory response directed towards H-2^b MHC class I molecules or derived peptides when expressed at high density on donor cells in an inflammatory microenvironment, such as occurs in transplantation.

Considering this hypothesis, it is relevant to analyse the T-cell repertoire in this model of transplantation in TAP1^–/– mice and to evaluate whether selected T-cell populations would be mobilized during rejection. There has been limited investigation of the diversity of the T-cell receptor (TCR) repertoire in TAP1^–/– mice, and few studies have reported the TCR V-beta (BV) diversity of the CD8⁺ T-cell population in these mice.^5,13 It was found that all BV subsets analysed in splenocytes were expressed at levels similar to those of wild-type animals (B6 and 129/Sv mice), with the exception of BV5·1/5·2 subsets in wild-type mice, which were poorly expressed.¹³

We measured the T-cell repertoire in these mice by analysing TCR BV–J-beta (BJ) usage and complementarity determining region (CDR3) length spectratyping, both before and after exposure to the graft or in response to peptides derived from the sequence of the H-2K^b class I molecule. The spectratyping technique was used to determine whether there was any preferential expansion of particular T cells, by contrasting CDR3 patterns of splenocytes in transplanted mice with those of naïve mice. We observed expanded T cells with limited BV-BJ expression during graft rejection in response to synthetic self class I H-2K^b peptides, as compared with Gaussian profiles in naïve animals. This study is the first analysis by CDR3 spectratyping of TCR BV-BJ chains expressed by TAP1^–/– mice during the inflammatory response against B6 grafts. Our data show the presence of oligoclonal expansions in the peripheral T-cell repertoire of transplanted TAP1^–/– mice in response to self H-2K^b peptides, indicating that autoreactive T-cell repertoires are indeed mobilized during rejection.

Materials and methods

Mice

TAP1^–/– mice³ were backcrossed to C57BL/6 (B6) mice 10 times before this study. B6 mice were purchased from the animal facility of the University of São Paulo, Brazil. All mice were used at 6–10 weeks of age and maintained in autoclaved microisolators in our own animal facility (Tropical Medicine Institute, University of São Paulo, São Paulo, Brazil). Animal care was in accordance with institutional ethical guidelines.

Peptides

Five synthetic (15-mer) peptides derived from H-2K^b MHC class I molecules were synthesized in our laboratory using published sequences.¹⁴ Peptides were selected based on reported motifs for I-Ab class II molecules.¹⁵ All peptides were purified by reverse-phase high-performance liquid chromatography (RP-HPLC) and analysed by mass spectrometry (Micromass/TofSpec SE, Micromass, Manchester, UK), and displayed >90% purity as determined by RP-HPLC analysis. The amino acid sequences of H-2K^b peptides are as follows:

• Peptide 1 (39–53 aa) GEPRYMEVGYVDDTE

• Peptide 2 (54–68 aa) FVRFDSDAENPRYEP

• Peptide 3 (69–83 aa) RARWMEQEGPEYWER

• Peptide 4 (84–98 aa) ETQKAKGNEQSFRVD

• Peptide 5 (103–117 aa) LGYYNQSKGGSHTIQ.

Non-vascularized heart transplantation

We used a modified model of heart transplantation.^16,17 B6 were donors to TAP1^–/– mice. One-day-old newborn donor hearts were implanted subcutaneously into the ear of male TAP1^–/– (H-2^b) mice. Rejection was confirmed by histological analysis.

Histology

Graft histology was performed on tissue sections (4 mm) after paraffin embedding and staining with haematoxylin and eosin, using standard protocols. Grafts were scored by a pathologist blinded to experimental groups. The heart graft tissues were classified into two groups: no rejection (no or minimal inflammation and no myocyte damage) or rejection (moderate to marked mononuclear infiltration, perivascular infiltration, myocyte damage, and occasional haemorrhage and fibrosis).

In vitro stimulation with H-2K^b peptides

Two pools of splenocytes from five transplanted and five naïve TAP1^–/– mice were obtained. The splenocytes were incubated with each of the five H-2K^b peptides (2·5 mg/ml) at 5 × 10⁶ cells/well (10⁶ cells/mice) for 3 days, in 48-well plates (Costar, Cambridge, MA) in serum-free HL-1 medium (Biowhittaker, Walkersville, MD), supplemented with 2 mm l-glutamine and 1 mm sodium pyruvate. Aliquots of each pool of cells were cultured under the same conditions without peptides. All cultures were incubated at 37° in a humidified atmosphere of 5% CO². After 3 days of culture, the cell suspensions were collected for RNA extraction. Splenocytes from these five transplanted TAP1^–/– mice were also incubated individually with each H-2K^b peptide (5 × 10⁶ cells/well/peptide).

Spectratyping analysis

Repertoire analysis was performed using a modified protocol described by Pannetier et al.¹⁸ We used a polymerase chain reaction (PCR)-based technology, in which fluorescent products which had been extended through the CDR3 region in each BV family were loaded onto an automated DNA sequencer and analysed with the GeneScan 2·0 software (Applied Biosystems, Foster City, CA).¹⁸ The length of the CDR3 region was determined by BV–C-beta (BC) segment amplification using a specific primer for the BC and different primers for each BV family. To analyse TCR BV usage in transplanted TAP1^–/– mice, we collected the splenocytes on day 10 or 20 post-transplantation, and the grafts were also harvested at this time for histopathological confirmation of rejection. Splenocytes from five transplanted TAP1^–/– mice were pooled and stimulated in vitro with five different H-2K^b synthetic peptides. After 72 hr, splenocytes were processed for RNA extraction. The same procedure was used for a pool of non-stimulated splenocytes from five naïve TAP1^–/– mice. Also, splenocytes from five transplanted TAP1^–/– mice were individually stimulated in vitro with H-2K^b peptides and used for CDR3 spectratyping. The repertoire in transplanted TAP1^–/– mice was compared with that found in the splenocytes from naïve TAP1^–/– animals.

Primary amplification of TCR BV families

Total RNA was extracted from splenocytes using Trizol reagent, according to the manufacturer's instructions (Gibco BRL Life Technologies, Grand Island, NY). RNA (1 mg) was reverse-transcribed into cDNA (12 ml) with an oligo(dT20) primer and SuperScript^TM II RT (Gibco BRL Life Technologies) transcriptase enzyme. Aliquots of cDNA (0·6 ml) were amplified by PCR using each BV-specific primer as the forward primer¹⁹ and a single BC primer as the reverse primer (BC145, 5′-CACTGATGTTCTGTGTGACA-3′). All primers used in this study were purchased from Operon Technologies (Alameda, CA); the sequences of the primers have previously been reported.²⁰ The PCR reaction was set up as follows: cDNA, 2·5 mm of each primer, 10 mm of BC145 primer, 10 mm of dNTP mix, 2·5 mm MgCl₂, and 1·25 U of Amplitaq DNA polymerase (AB Applied Biosystems, Foster City, CA) in 10 × PCR Buffer II, AB Applied Biosystems, Foster City, CA. Amplifications were conducted in GeneAmp PCR system 9700 (AB Applied Biosystems). The reactions started with a 5-min denaturation step at 94°, followed by 39 cycles of 45 seconds at 94°, 45 seconds at 55°, and 45 seconds at 72°, and ended with an elongation step of 10 min at 72°.

Secondary amplification of BV-BJ gene segments

Using 2 ml of the PCR product of BV-BC as a template, run-off reactions were performed with each of the 12 BJ fluorescent primers. The PCR reaction was set up as follows: 2 ml of PCR product, 1 mm of each BJ primer, 10 mm of BC145 primer, 10 mm of dNTP mix, 2·5 mm MgCl₂, and 0·5 U of Taq-E (AB Applied Biosystems) in 10 × PCR Buffer II. Amplifications were conducted in GeneAmp PCR system 9700 (AB Applied Biosystems). The reactions started with a 5-min denaturation step at 94°, followed by 20 cycles of 45 seconds at 94°, 45 seconds at 55°, and 1 min at 72°, and ended with an elongation step of 5 min at 72°. After run-off extensions with 12 BJ fluorescent primers, the run-off products were separated on the basis of their length on the automated 310 Genetic Analyser ABI PRISM with POP-4 polymer (PE Applied Biosystems). The results were analysed using GeneScan 2·0 software.

The relative index of stimulation (RIS) value was calculated as the area under the experimental peak divided by the area under the control peak found within a Gaussian distribution. Control peaks were obtained from naïve TAP1^–/– mice. A RIS value ≥4 for a particular peak was considered a significant expansion of T cells with that particular CDR3 length.

Results

Polyclonal T-cell repertoire at BV level in the periphery of transplanted TAP1^–/– recipients

We analysed alterations of the TCR repertoires at the BV-BC level in splenocytes from two groups of animals: TAP1^–/– mice that had rejected heart grafts from B6 donors (n = 5) and naïve TAP1^–/– mice (n = 5).

Comparing transplanted and naïve TAP1^–/– mice, we observed no differences in the spectratyping analysis of the CDR3 size distribution of BV families. Both groups of TAP1^–/– mice showed rearrangements of almost all BV families, which have also been reported to be expressed in B6 mice, which have the same H-2 background (data not shown).^21–23 One exception was the BV5·3 transcript, which could not be amplified from either group of TAP1^–/– mice. The TCR β-chain CDR3 spectratyping showed typical polyclonal T-cell spectra, following a Gaussian distribution.

Significant expansions at the CDR3 level after in vitro stimulation with H-2K^b peptides in splenocytes from TAP1^–/– mice that rejected B6 grafts

BV-BC analysis provided a global picture of the peripheral repertoire in TAP1^–/– mice during the rejection of B6 grafts, but no evidence regarding particular clonal expansions could be obtained from such a large group of clones. To obtain specific and detailed information about possible clonal expansions within the T-cell repertoire, all of the BV-BC rearrangements were elongated with the 12 labelled BJ primers. To identify dominant TCR rearrangements within a specific BV-BJ combination, we compared the BV-BJ profiles for transplanted TAP1^–/– mice after in vitro stimulation by individual H-2K^b peptides with the profiles for naïve TAP1^–/– mice. The significance of these expansions was assessed from the RIS value (see Materials and methods).

As summarized in Table 1, the BV-BJ analysis of a pool of splenocytes from transplanted TAP1^–/– mice stimulated in vitro with H-2K^b peptides showed 20 significant expansions (RIS value ≥4) that were not restricted to a specific Kb-derived peptide. Importantly, nine out of 20 (45%) significant BV-BJ expansions were also detected ex vivo in transplanted TAP1^–/– mice, suggesting that transplantation itself may be responsible for these expansions (Table 1).

Table 1.

Dominant V-beta–J-beta (BV-BJ) expansion detected in transplanted TAP1^−/− mice after in vitro stimulation with self H-2K^b peptides

		Peptide 1		Peptide 2		Peptide 3		Peptide 4		Peptide 5
BV-BJ		RIS	CDR3*	RIS	CDR3†	RIS	CDR3	RIS	CDR3	RIS	CDR3
BV3S1	BJ1S5	8	7	Not tested		4	7	10	7	<2
BV4	BJ1S3	<2		<2		9.7	11	<2		10	11
BV5S1	BJ1S1†	5	7	<2		4.6	6	<2		5.2	9
	BJ1S2†	<2		<2		4.2	7	4.6	7	Not valid
	BJ1S5†	4.3	8	11	8	Not valid		14	8	Not valid
	BJ2S1†	<2		<2		<2		4	10	Not tested
	BJ2S2	10	6	Not tested		9.8	6	4	9	<2
	BJ2S3	8	10	11	10	9.8	10	5	10	<2
BV6	BJ1S2	<2		<2		<2		4.8	7	Not valid
	BJ2S4	5.7	5	<2		<2		<2		Not tested
	BJ2S7	<2		<2		<2		15	4	14	4
BV8S1	BJ1S5†	<2		6.2	8	5.2	8	<2		<2
BV11	BJ1S2	<2		<2		4.9	6	<2		<2
	BJ1S3†	5.2	9	4.4	9	4.7	9	<2		<2
	BJ1S6	Not tested		<2		4.5	8	<2		<2
	BJ2S1‡	4.8	10	4	10	<2		<2		<2
BV12	BJ1S4‡	<2		4	5	<2		4	5	<2
BV14	BJ1S5	7.4	7	Not tested		<2		Not tested		<2
	BJ2S4	9.7	6	Not tested		4.4	6	Not tested		<2
BV16	BJ1S3	9	7	9.6	7	8.4	7	11	7	5.3	7

Bold font represents significant BV-BJ expansions induced by H-2K^b peptides which showed oligoclonal profiles (single peaks).

Reverse-transcribed RNA from cultured splenocytes from a pool of five transplanted TAP1^–/– mice was amplified with BV- and BJ-specific primers. Run-off products were then analysed in a DNA sequencer.

CDR, complementarity determining region; RIS, relative index of stimulation.

^*The RIS value was calculated as the area under the experimental peak divided by the area under the control peak found within a Gaussian distribution in naïve TAP1^–/– mice. Significant expansion was defined with RIS value ≥4.

^†CDR3 length, in amino acids, of expanded peak.

^‡BV-BJ rearrangements that exhibited dominant expansions in transplanted TAP1^–/– mice ex vivo, without peptide stimulation.

Most interestingly, the T-cell repertoire in transplanted mice showed remarkable differences when compared with non-stimulated peripheral lymphocytes in naïve TAP1^–/– mice. Unlike the multiple-peak pattern with Gaussian distributions in naïve mice, one dominant peak representing a single-sized CDR3 fragment emerged above the background of multiple peaks in several BV-BJ gene products. These dominant peaks, which sometime were single peaks, were found in the rearrangements BV3S1/BJ1S5 (CDR3 7 aa), BV5S1/BJ1S5 (CDR3 8 aa), BV6/BJ1S2 (CDR3 7 aa), BV8S1/BJ1S5 (CDR3 8 aa), BV11/BJ1S3 and BJ1S6 (CDR3 9 aa and 8 aa), BV12/BJ1S4 (CDR3 5 aa), BV14/BJ1S5 and BJ2S4 (CDR3 7 aa and 6 aa), and BV16/BJ1S3 (CDR3 7 aa), after in vitro stimulation with different H-2K^b peptides. The patterns of some of these rearrangements are shown as representative examples in Fig. 1. The other BV-BJ rearrangements with significant peaks (RIS ≥4) (shown in Table 1) are less dominant expansions of T-cell populations which were poorly represented in naïve TAP1^–/– mice.

Figure 1.

BV5S1/BJ1S5 BV8S1/BJ1S5 BV11/BJ1.3 BV12/BJ1S4 BV14/BJ2S4 BV16/BJ1.3 Dominant V-beta–J-beta (BV-BJ) expansions in transplanted transporter associated with antigen processing 1 (TAP1)^–/– mice after 3 days of in vitro stimulation with H-2K^b peptides. All of the BV-BJ rearrangements were determined on a splenocyte pool from five different transplanted TAP1^–/– mice collected at the time of rejection, and stimulated in vitro with H-2K^b peptides, in comparison with a splenocyte pool from naïve TAP1^–/– mice without stimulation in vitro. BV-BJ profiles of six different rearrangements, most displaying oligoclonal peaks, are shown in transplanted TAP1^–/– mice. For comparison, at the bottom of each BV-BJ rearrangement are the immunoscope profiles of naïve TAP1^–/– mice. Total RNA was extracted from splenocyte pools, and cDNA was synthesized and amplified with primers specific for BV families. Run-off reactions were then performed using the BJ-specific fluorescent primers, followed by β-chain spectratyping. The areas of each peak were obtained with the GenScan 2·0 software. The intensity of fluorescence is represented in arbitrary units (y-axis) as a function of the complementarity determining region 3 (CDR3) length in base pairs (x-axis). The number shown under the peaks corresponds to the CDR3 length (aa) for each particular expansion. H-2K^b peptide sequences are shown in Materials and methods.

Detection of a public T-cell expansion in transplanted mice ex vivo indicates occurrence of in vivo phenomenon

To investigate if these dominant expansions of T cells were shared by different recipients, splenocytes were harvested at the time of rejection (between the 10th and 20th day post-transplantation) in five individual TAP1^–/– mice that had rejected B6 heart grafts. At that time, cells were stimulated in vitro with each individual H-2K^b peptide. The BV-BJ rearrangements that showed significant expansions or altered patterns in the analysis of splenocytes pooled from transplanted TAP1^–/– mice after in vitro stimulation with H-2K^b peptides were analysed individually. We were not able to study all animals in each condition because of an insufficient quantity of cDNA.

A summary of most of the selected BV-BJ expansions, showing the restricted CDR3 length of peripheral cells from individual animals, is presented in Table 2. Eight different BV-BJ expansions were identified in the majority of the studied animals in the response to at least one H-2K^b peptide. These expansions showed a restricted CDR3 length as in the analysis of the splenocyte pool from transplanted TAP1^–/– mice after in vitro stimulation. It should be noted that some of these expansions (BV5S1/BJ1S1, BV5S1/BJ1S2, BV5S1/BJ1S5, BV5S1/BJ2S4 and BV11/BJ1S3) with restricted CDR3 length were also observed ex vivo in splenocytes from transplanted TAP1^–/– mice without in vitro stimulation with H-2K^b peptides, indicating in vivo expansion following transplantation. Moreover, most of the dominant expansions (BV5S1/BJ1S2, BV5S1/BJ2S4, BV6/BJ1S2, BV9/BJ2S1 and BV11/BJ1S3) were detected in most of the animals studied (Fig. 2).

Figure 2.

T-cell clonal dominance in peripheral cells from several transplanted transporter associated with antigen processing 1 (TAP1)^–/– mice. All of the V-beta–J-beta (BV-BJ) spectratypes were determined on splenocytes from five different transplanted animals collected at the time of rejection. The spectratyping profiles for four out of five BV-BJ rearrangements observed in all transplanted TAP1^–/– mice in response to at least one peptide are shown. For representation, we selected only those histograms of BV-BJ subfamilies with restricted complementarity determining region 3 (CDR3) profiles after stimulation with a specific H-2K^b peptide: BV11/BJ1S3 after stimulation with peptide 3, BV5S1/BJ1S2, BV5S1/BJ2S4 and BV6/BJ1S2 after stimulation with peptide 4. The most restricted profiles are boxed. For comparison, at the top of each BV-BJ rearrangement are the spectratyping profiles of naïve TAP1^–/– mice and transplanted animals without in vitro stimulation with H-2K^b peptides. Graphs represent the intensity of fluorescence in arbitrary units as a function of run-off product size in nucleotides. The length of each public expansion is indicated in base pairs at the bottom of each rearrangement. Animal numbers are given on the left.

Despite the highly selected T-cell repertoire responsive to the set of self class I peptides, not all of these BV-BJ expansions showed the same spectratyping profile (Table 2). The BV5S1/BJ1S2, BV5S1/BJ1S5, BV5S1/BJ2S4 and BV6/BJ1S2 rearrangements showed mostly single peaks, in contrast with the other four rearrangements which exhibited significant peaks over a polyclonal background.

Table 2.

Oligoclonal public T-cell expansions and polyclonal positive expansions in the peripheral repertoire from individual transplanted TAP1^–/– mice^*

				Mice with positive expansions after in vitro stimulation with H-2Kb peptides
				Mice tested with each peptide
BV-BJ rearrangements (CDR3 size)	Mice with positive expansions†	Mice with oligoclonal expansions‡	Mice with polyclonal expansions§	1	2	3	4	5	Profile
BV5S1/BJ1S1¶	4/5	0/4	4/4	0/3	0/2	0/3	ND	0/5	O
(7 aa)	80%	0%	100%		1/2			3/5	P
BV5S1/BJ1S2¶	5/5	3/5	2/5	0/3	0/4	3/5	1/5	1/5	O
(7 aa)	100%	60%	40%		2/4	1/5	3/5	3/5	P
BV5S1/BJ1S5§	3/5	2/3	1/3	ND	1/2	0/4	1/5	2/5	O
(8 aa)	60%	67%	33%		0/2	1/4	1/5	0/5	P
BV5S1/BJ2S4¶	5/5	5/5	0/5	1/3	4/4	2/4	5/5	2/4	O
(8 aa)	100%	100%	0%	2/3	0/4	2/4	0/5	0/4	P
BV6/BJ1S2	5/5	3/5	2/5	0/5	ND	ND	3/5	ND	O
(6 aa)	100%	60%	40%	2/5			2/5		P
BV6/BJ2S4	3/4	0/3	3/3	0/4	ND	ND	ND	ND	O
(5 aa)	75%	0%	100%	3/4					P
BV9/BJ2S1	5/5	1/5	4/5	1/5	ND	ND	ND	1/5	O
(4 aa)	100%	20%	80%	1/5				3/5	P
BV11/BJ1S3¶	5/5	1/5	4/5	1/4	0/2	1/5	ND	ND	O
(9 aa)	100%	20%	80%	1/4	2/2	1/5			P

Bold font indicates dominant oligoclonal expansions.

The ratios shown are the number of mice with positive expansions in relation to the number of studied animals.

^*Significant BV-BJ expansions detected in the pool from transplanted animals were also analysed in five transplanted TAP1^–/– mice. Note that not all animals were studied for all peptides, because of limitations of the biological material.

^†For this analysis, we considered the animals with positive expansion in response to at least one H-2K^b peptide.

^‡For this analysis, we considered the animals with dominant oligoclonal expansions in response to at least one H-2K^b peptide.

^§For this analysis, we considered the animals with positive expansions inside a polyclonal profile in response to at least one H-2K^b peptide.

^¶Oligoclonal expansions present in transplanted TAP1^–/– mice ex vivo without in vitro stimulation with peptides.BV-BJ, V-beta–J-beta; CDR, complementarity determining region; ND, not done; O, oligoclonal profiles; P, polyclonal profiles.

In addition, the BV5S1/BJ2S4 dominant expansion (177 bp, CDR3 length 8 aa) was single-peak in all animals tested in response to at least one peptide (Fig. 3) and also exhibited the same pattern in transplanted TAP1^–/– mice without in vitro peptide stimulation. Thus, we defined it as a public expansion (Table 2). Nonetheless, the analysis of the TCR repertoire by TCR CDR3 length distribution does not take into account the amino acid differences in the CDR3 and the possibility that one or several T-cell clones may be contained within a peak at a particular CDR3 length.

Figure 3.

BV5S1/BJ2S4 public expansion in transplanted transporter associated with antigen processing 1 (TAP1)^−/− mice. Reversetranscribed RNA of splenocytes from transplanted TAP1^−/− mice was amplified with BV5S1- and TCR-BC-specific primers. Run-off reactions were then performed using the BJ2S4 fluorescent primer. Analysis of run-off products was assessed as described in Fig. 1. Shown are BV5S1/BJ2S4 profiles in spleen cells from naïve TAP1^−/− mice, from transplanted TAP1^−/− recipients without stimulation with peptides, and from different animals after 3 days of in vitro stimulation with H-2Kb peptides. Graphs represent the intensity of fluorescence in arbitrary units as a function of run-off product size in nucleotides. Animal numbers are given on the left.

It was interesting to note that the BV5S1/BJ2S4 rearrangement with the same CDR3 length (8 aa) was observed in all recipients stimulated with four different peptides (peptides 1, 2, 3 and 4) (Table 2). However, this public expansion only exhibited single peaks in all recipients after in vitro stimulation with peptides 2 and 4, as shown in Fig. 3.

It was of interest to evaluate whether the H-2K^b peptides showed differences in relation to the dominant expansions that these peptides induced. As shown in Table 2, we did not find one particularly immunodominant peptide. Surprisingly, it seems that one repertoire family, BV5S1/BJ2S4, predominates. In the present study, BV-BJ rearrangements with different profiles exhibited expansions in transplanted TAP1^–/– mice in response to at least one of the H-Kb peptides.

Discussion

Our study involves a model of transplantation which includes a strong inflammatory response similar to the ones directed to alloantigens, but the antigens involved in triggering rejection are in fact self antigens. It can therefore be considered to be a model of inflammatory autoimmunity.

TCR spectratyping was used to analyse the TCR repertoire of peripheral T cells from TAP1^–/– mice involved in the response against synthetic peptides derived from the H-2K^b sequence which we had previously shown to be a molecular target in the rejection of B6 grafts.¹¹ Analysis by examination of all of the TCR BV genes revealed that T cells from transplanted TAP1^–/– mice, collected during the rejection of B6 heart grafts, used a wide spectrum of TCR BV genes with a Gaussian distribution. Nevertheless, a finer analysis involving the 12 BJ families for each BV showed important and restricted CDR3 expansions at the BV-BJ level. These results suggest the existence of significant expansions of clones with several different BV-BJ rearrangements in response to these class I-derived peptides during rejection, and that distinct T-cell populations were expanded. Some of these T-cell expansions were found after rearrangement of a specific BV-BJ combination, and others after rearrangement of a specific BV family but with different BJ segments.

These findings are consistent with our hypothesis that TAP1^–/– mice may present either defective control of the autoreactive T-cell repertoire or functional alterations in their autoreactive CD4⁺ T-cell repertoire, which expands and mounts an aggressive response to tissues with a relatively high density of self MHC class I molecules in an inflammatory environment. Although all the TCR spectratyping was performed on unseparated total splenic T cells, the expanded T cells are probably CD4 T cells, as CD8⁺ T cells were only a minor population in these TAP1^–/– mice (mean 2%) and our previous study¹¹ showed that CD4 T cells play an important role in this model of rejection. Nevertheless, further studies are being carried out to clarify whether expansion of residual CD8⁺ T cells might influence CDR3 length spectratyping analysis.

We believe that the inflammatory milieu contributes to the activation of these potentially aggressive T cells which escape peripheral regulatory mechanisms, probably impaired in TAP1^–/– mice. It is relevant to mention that, although B6 mice have a high frequency of T-cell precursors reactive to these peptides and show a higher proliferative response than naïve TAP1^–/– mice,²⁴ they do not developed spontaneous autoimmune disease. In contrast, TAP1^–/– mice develop spontaneous autoimmune arthritis when housed under conventional conditions, suggesting that these mice are unable to efficiently activate regulatory mechanisms and maintain homeostasis.²⁵ We suggest that the absence in the periphery of certain regulatory CD8⁺ T-cell populations in these mice may contribute to a lack of control over these potentially inflammatory self-reactive CD4 T cells, as reported in other models.^26–32

In the present study, the highly selected repertoire at the BV-BJ level was more evident after in vitro stimulation with self H-2K^b peptides, and 55% of these expansions showed single peaks (see Table 1). These expansions within the CDR3 profiles showing restricted BV-BJ usage possibly contain activated T cells, including autoreactive T-cell clones involved in the inflammatory response against H-2^b grafts. Nevertheless, these expansions may also contain T cells with regulatory function, which are also mobilized in inflammatory processes.

We detected one public oligoclonal expansion and other dominant expansions in transplanted TAP1^–/– mice in response to self H-2K^b peptides. This result is in accordance with other data both in rejection and in autoimmune diseases, showing a limited number of public clones,³³ and supports our hypothesis that the aggressive response against a B6 graft developed by TAP1^–/– (H-2^b) mice was probably triggered by a process analogous to pathological autoimmunity. In this regard, it is relevant to compare related data on both autoimmune diseases and allotransplantation.

In other models of allotransplantation in wild-type animals, data concerning CDR3 spectratyping are very heterogeneous, showing both restricted and polyclonal BV T-cell repertoires, during rejection and tolerance.^34–38

Several investigators have looked at autoantigen-specific T-cell populations in autoimmune diseases in a variety of analyses of the T-cell repertoire.^39–42 It has been reported that the T-cell response to self antigens can be directed to different epitopes or focused on a few immunodominant epitopes. In a model of experimental autoimmune encephalomyelitis (EAE) in B10.PL mice, it was shown that a majority of myelin basic protein (MBP)-reactive CD4⁺ T cells recognized an immunodominant peptide and that these dominant pathogenic public T-cell clones used the TCR BV8S2/BJ2S7.⁴¹ The apoptotic depletion of these BV8S2 T cells, mediated by BV14⁺ regulatory CD4⁺ T cells and regulatory CD8⁺ T cells, resulted in remission from EAE, confirming the pathogenic activity of this clonal population.⁴³ However, oligoclonal expansions have also been reported in healthy individuals and are not exclusively a feature of pathological conditions.^18,44,45

An important issue to consider is whether the animals develop a shared dominant response to H-2^b class I-derived peptides (public response) or whether each animal can develop its own particular response (private response). In animal models of autoimmune diseases, public clones, found in each affected animal, would favour the notion that certain clones are more closely connected to disease than others. Our study has revealed single-peaked dominance within particular BV-BJ combinations after in vitro stimulation with H-2K^b peptides in individual transplanted TAP1^–/– mice. Interestingly, one of these peaks (in the family BV5S1/BJ2S4) was public, suggesting the occurrence of particular oligoclonal T-cell populations shared by different animals that expanded in response to the graft. We favour the idea that, during the initial phase of rejection of grafts with a relatively high density of MHC class I, multiple H-2^b-derived peptides shed from the graft are processed and presented by self antigen presenting cells (APCs) inducing the mobilization of a diverse T-cell repertoire. As rejection progresses, T-cell repertoires, possibly comprising T cells with higher avidity for H-2^b class I peptides, would persist, displaying a more restricted TCR usage. Thus, we have shown that naïve TAP1^–/– mice activate T cells reactive to self H-2K^b peptides as part of their peripheral T-cell repertoire; however, transplanted TAP1^–/– mice displayed a much higher T-cell response to these peptides than naïve TAP1^–/– mice.²⁴

It is important to emphasize that the public oligoclonal BV5S1/BJ2S4 expansion which occurred in vivo during the rejection of B6 heart grafts by TAP1^–/– mice was also detected in individual transplanted TAP1^–/– mice, after in vitro stimulation with different H-2K^b peptides. This supports the idea that the BV5S1/BJ2S4 public expansion was indeed autoreactive and possibly degenerate in specificity to a diverse set of peptides derived from self class I molecules. More likely is the proposition that certain CDR3 lengths, coupled to particular BVBJ families, are strongly favoured and selected in response to particular antigenic determinants. The identification of such expanded T-cell clones may allow the development of strategies to manipulate putative inflammatory T-cell populations, as described in other systems.^30,46–48 Such defined T-cell expansions with a broad pattern of peptide recognition may be characteristic of dominant T-cell populations, preferentially recruited in TAP1^–/– mice during the aggressive response to B6 grafts.

It should be pointed out, however, that functionally distinct T-cell subpopulations, such as regulatory and effector T cells, and the microenvironment may contribute to the recruited restricted profiles. As we analysed the T-cell repertoire in the periphery and not directly in rejected grafts, where graft-infiltrating lymphocytes could use a TCR repertoire distinct from that of the peripheral T-cell pool,⁴⁹ it is important to compare the two repertoires to determine whether these oligoclonal T-cell expansions are also present in the rejected grafts. Experiments are in progress to test this possibility.

Although the functional activities of these T-cell populations were not explored in this study, the identification of dominant and public reactive clones may be relevant for the induction of transplantation tolerance, either by the elimination of effector T cells or by the stimulatiion of regulatory T cells, or both. Based on this idea, it has been shown that, in the B10.PL mouse model of EAE, the in vivo expansion of specific CD4⁺ and CD8⁺ regulatory T-cell populations following immunization with BV8S2 TCR peptides, recombinant single-chain TCR BV8S2 proteins, BV8S2 plasmid DNA, or recombinant adenovirus vector expressing TCR BV8S2 protected mice from developing disease.^50,51

In conclusion, this study demonstrated the expansion of a small number of T-cell clones during the rejection of B6 grafts as part of the peripheral T-cell repertoire in transplanted TAP1^–/– mice. The oligoclonality in the analysed T-cell repertoire, probably of CD4⁺ T cells, is likely to reflect an autoreactive process developed by TAP1^–/– mice. The appearance of these oligoclonal expanded T cells in transplanted TAP1^–/– mice emphasizes the need for further analysis of T-cell clonality in graft-infiltrating lymphocytes. Further studies will be required to establish the functionality of these expanded T cells, and should allow us to explore the CDR3 sequences of members of the T-cell repertoire in TAP1^–/– mice. Also, impaired immune regulatory mechanisms that may be involved in the related pathological autoreactivity in this model should be examined.

These results present a model of inflammatory autoimmunity in which a limited T-cell repertoire is used for the immune response to this graft. Just as in experimental models of autoimmune diseases, driver clones can possibly be sought and found, which may propel and propagate the transplant response. It seems likely that, both in pathological autoimmunity and in transplantation, the more narrow the T-cell repertoire involved in the inflammatory response, the easier it should be to devise control regulatory methodologies to protect the individual from tissue aggression. Important issues to be addressed are whether it is worthwhile to define a set of public drivers crucial for the aggressive response, and to design specific therapies against this driver population, or rather, to activate and expand large numbers of regulatory T cell to most efficiently control transplant reactivity. This choice may be more relevant in the context of human transplantation, in which the alloantigen repertoire capable of triggering rejection is very diverse even in the early phase of the inflammatory process.

Abbreviations

APC

antigen presenting cell

BC

C-beta

BJ

J-beta

BV

V-beta

CDR

complementarity determining region

CFA

complete Freud's adjuvant

EAE

experimental autoimmune encephalomyelitis

MBP

myelin basic protein

MHC I

major histocompatibility complex class I

NK

natural killer

PCR

polymerase chain reaction

RIS

relative index of stimulation

TAP

transporter associated with antigen processing

TCR

T-cell receptor