It is well established that melanoma cells express Ags that are recognized by autologous T cells in vitro. Tumor-infiltrating lymphocytes in situ comprise clonotypic T cells, suggesting that their expansion is driven by Ag stimulation. Still, little is known about the detailed characteristics of the in situ T cell response. In the present study, we scrutinized this response by analyzing multiple metastatic lesions for the presence of clonotypic T cells. This approach was chosen to distinguish whether the clonal T cell expansion occurs as a systemic or localized phenomenon. TCR clonotype mapping of six s.c. metastases from two patients revealed the presence of multiple (from 40 to >60) clonotypic T cells in all lesions. Clonotypic T cells were present in TCR β-variable regions expressed both at high and low levels. Comparison of the T cell clonotypes present in different lesions from individual patients demonstrated that, in general, clonotypes were exclusively detected in a single lesion. Hence, anti-melanoma T cell responses are much more heterogeneous than previously anticipated and accommodate a predominance of strictly localized T cell clonotypes.

It has been shown that the immune response to melanoma includes CTLs capable of lysing autologous melanoma cells in vitro. Several melanoma-associated Ags (MAA)3 have been characterized, and the peptide epitopes have been defined (1). Analyses of tumor-infiltrating lymphocytes (TILs) in situ have demonstrated the presence of clonotypic T cells within the tumor, strongly indicating that these T cells recognize MAA in the context of HLA molecules (2). However, most studies have been limited to the analysis of T cell clonality in subpopulations of TILs expressing specific TCR V regions at a significant level (2, 3). Thus, the complexity and heterogeneity of the T cell clonotypes engaged with the responses against melanoma may have been underestimated.

Tumor-specific CTLs are present in the blood of melanoma patients. However, it remains unknown whether these cells actually take part in the immune response against melanoma and in more general terms whether the T cell response against melanoma occurs as a systemic or localized phenomenon. In this respect, powerful antitumor reactivity in the PBL in most cases does not correlate with tumor regression. The background for this enigma is not known, but it could possibly be related to the failure of melanoma-specific CTLs to localize to the tumor site(s).

In the present study, we have used a recently developed, highly sensitive technique based on RT-PCR and denaturing gradient gel electrophoresis (DGGE) to analyze for the presence of clonotypic T cells in six s.c. melanoma lesions from two patients. Our results demonstrate that the T cell infiltrate in melanoma is exceedingly heterogeneous and accommodates a much higher number of T cell clonotypes than previously anticipated. Semiquantitative RT-PCR analyses revealed that T cell clonotypes are present not only in overexpressed β-variable (BV) regions, but also in (BV) regions expressed at low levels. Furthermore, in most cases specific T cell clonotypes were detected in a single lesion only, indicating that locally expanded clonotypic T cells dominate the T cell infiltrate in melanoma, and that these cells enter the periphery and home to other tumor sites only to a very limited extent.

Fresh s.c. melanoma lesions were received as biopsy material immediately after surgery performed at the Department of Plastic Surgery (University Hospital, Copenhagen, Denmark) (patient 1) or the Department of Dermatology (School of Medicine, Würzburg, Germany) (patient 2). Three lesions were excised from each of the patients from the following anatomical sites: Patient 1, right thorax (all lesions); patient 2: axilla (lesion 1), axilla (lesion 2), and forehead (lesion 3). Tumor lesions and blood samples were taken from the respective patients at the same timepoint. PBLs were used for tissue typing (patient 1: HLA-A1, A2, B8, B44, Cw3, Cw5, DR3, DR9; patient 2: HLA-A2, B15, B38; typing was only performed for HLA class I A and B molecules in this patient).

Frozen sections were fixed in cold acetone for 10 min followed by removal of endogenous peroxidase with 0.03% H2O2 and blocking of collagenous elements with 10% species-specific serum in 1% BSA (Boehringer Mannheim, Mannheim, Germany)/PBS. Next, the Abs were overlayed onto serial sections at predetermined dilutions (usually 20 μg/ml), and the slides were incubated in a humid chamber for 30 min. With PBS washes between every step, a biotinylated link Ab was applied for 10 min followed by a streptavidin-linked enzyme (i.e., either peroxidase or alkaline phosphatase) for 10 min. Substrate was added after another wash, and the slides were incubated in the dark for 20 min. After a wash in PBS, the slides were counter stained, mounted, and viewed using an Olympus BH2 microscope (Hamburg, Germany) with photographic capabilities.

Tumor biopsies and blood samples were processed immediately after surgery. RNA was extracted using the Purescript Isolation Kit (Gentra Systems, Minneapolis, MN). cDNA synthesis and quantitation of TCR cDNA in the each sample were conducted as described previously (4). In brief, 2–5 μg of RNA were used in the synthesis of cDNA using SuperScript II reverse transcriptase (Life Technologies, Gaithersburg, MD) in a total volume of 50 μl 1× buffer (Life Technologies) containing 10 mM DTT. The RT reactions were primed with a mixture of oligo(dT) and random hexamers, and incubations were performed at 37°C for 30 min, 42°C for 30 min, and 72°C for 5 min. Using equal amounts of TCR template in all reactions, cDNA was amplified using separate primer panels for quantitative TCRBV region analysis (4) and TCR clonotype mapping by DGGE (5). The sequences of these primer panels are available upon request. Amplifications were conducted in a total volume of 25 μl containing 1× PCR buffer (50 mM KCl, 20 mM Tris (pH 8.4), 2.0 mM MgCl2, 0.2 mM cresol red, 12% sucrose, and 0.005% (w/v) BSA (Boehringer Mannheim)), 2.5 pmol of each primer, 40 mM of dNTP (Pharmacia LKB, Uppsala, Sweden), and 1.25 U of AmpliTaq polymerase (Perkin-Elmer Cetus, Emeryville, CA). The parameters used for amplification were 94°C for 60 s, 60°C for 60 s, and 72°C for 60 s for 32 (quantitative PCR) or 40 cycles (DGGE). Taq polymerase and dNTPs were added to the reaction tube at an 80°C step between the denaturation and annealing steps of the first cycle (Hot start).

For quantitative analysis, the constant region primer was end-labeled with 33P. Aliquots (10 μl) of PCR products were electrophoresed in a 2% NuSieve 3:1 agarose gel (FMC BioProducts, Rockland, ME) that was subsequently dried under vacuum and exposed to a Storage Phosphor Screen (Molecular Dynamics, Sunnyvale, CA). Quantitation was accomplished using ImageQuant software (6) and expressed as a mean percentage (±SD) of the sum of the total TCRBV signal detected.

For DGGE analysis,10-μl aliquots were loaded onto a denaturing gradient gel containing 6% polyacrylamide and a gradient of urea and formamide from 20% to 80%. Gels were run at 160 V for 4.5 h in 1× TAE buffer (0.04 M Tris-acetate, 0.001 M EDTA) kept at a constant temperature of 58°C. After electrophoresis, the gel was stained with ethidium bromide and photographed under UV transillumination.

To analyze for the expression of tyrosinase, MART-1/Melan-A and glycoprotein 100 (gp100), RT-PCR was conducted with primers specific for these transcripts as described previously (7).

DNA bands that resolved at identical positions in the denaturing gradient gel were subjected to sequence analysis using the Thermo Sequenase cycle sequencing kit (Amersham Life Science, Cleveland, OH). In brief, bands were excised from the denaturing gradient gel, and DNA was eluted in H2O and reamplified. An aliquot (0.2 μl) of the PCR product was used as template in a 40-cycle sequencing reaction with 33P end-labeled β-constant region sequencing primer. Gels were dried under vacuum and exposed to a Storage Phosphor Screen.

PBLs and three s.c. melanoma lesions from each of the two patients were analyzed for the expression of TCRBV regions 1–24 by semiquantitative RT-PCR. In each tumor sample, at least two BV regions were expressed at significantly higher levels than in the peripheral blood (Table I). In patient 1, this applied to BV2, BV4, BV8, BV14, and BV23; in patient 2, this applied to BV3, BV5, BV8, BV19, BV20, and BV24. In patient 2, the BV regions BV3, BV20, and BV24 were highly expressed in two lesions, and BV8 was highly expressed in all three lesions. Major differences were detected at the level of BV region expression between the different lesions from each patient (e.g., patient 1, BV6, 2.8% in lesion 1 vs 10.1% in lesion 2; patient 2, BV3, 10.6% in lesion 2 vs 3.4% in lesion 3). Thus, there were not only differences in the relative expression of TCRBV regions in TILs and PBLs, but also between the different melanoma lesions.

Table I.

TCRBV region usage, determined by quantitative RT-PCR, and the number of T cell clonotypes, determined by TCR clonotype mapping, in s.c. melanoma lesions from patients 1 and 2a

TCRBVPatient 1Patient 2
Lesion 1Lesion 2Lesion 3PBLsLesion 1Lesion 2Lesion 3PBLs
Exp. %ClonesExp. %ClonesExp. %ClonesExp. %Exp. %ClonesExp. %ClonesExp. %ClonesExp. %
3.7 5.7 4.6 >4 4.1 (0.2) 3.5 (1.0) 8.4 (1.3) 8.1 (0.0) 9.2 (1.1) 
4.1 7.9 >4 3.0 3.3 (0.5) 2.7 (0.0) 3 5.4 (1.6) 1 5.0 (0.3) 3 5.9 (1.5) 
6.4 7.4 5.2 >4 7.0 (0.0) 6.6 (1.4) >4 10.6 (0.4) 3.4 (0.3) 3.1 (1.6) 
5.4 >4 9.8 2.3 3.3 (0.7) 6.4 (0.2) 4.0 (0.0) 7.9 (1.1) 5.7 (0.4) 
4.0 9.3 4.5 3.8 (0.1) 9.0 (1.6) >4 7.6 (2.8) 3.9 (0.0) 4.6 (2.2) 
2.8 10.1 6.0 4.2 (1.7) 4.7 (0.1) 4.0 (1.0) 6.9 (0.1) 3.9 (0.2) 
<1 <1 <1 1.5 (0.5) <1 (0.0) <1 (0.1) <1 (0.1) 4.6 (0.6) 
6.7 >4 9.5 2.9 >4 6.2 (0.7) 8.1 (0.5) 9.1 (0.7) 4 9.0 (2.1) 4 3.3 (0.4) 
3.5 2.9 3.1 4.0 (0.3) 2.5 (0.0) 6.3 (2.5) 1.8 (0.1) 4.4 (0.5) 
10 5.6 <1 3.5 4.0 (0.6) 7.3 (0.2) <1 (0.1) 1.4 (0.1) 3.7 (0.4) 
11 <1 <1 3.0 2.9 (0.6) 6.7 (1.4) <1 (0.1) <1 (0.1) 2.6 (0.8) 
12 3.8 <1 3.5 4.8 (0.8) 4.5 (1.6) >4 <1 (0.2) 7.2 (1.7) 3.9 (1.6) 
13 4.2 >4 3.8 2.7 >4 4.3 (1.9) 4.2 (1.2) >4 7.3 (0.2) <4 5.1 (0.7) 3 7.5 (0.7) 
14 6.3 >4 1.1 3.8 3.7 (0.3) 3.9 (1.1) >4 2.5 (0.0) 2 3.6 (1.0) 2 3.4 (0.8) 
15 1.5 <1 1.5 >4 2.5 (0.2) <1 (0.1) 1.5 (0.1) <1 (0.6) 1.5 (0.8) 
16 3.8 4.3 4.6 >4 3.2 (1.5) 2.6 (0.0) 1.0 (0.3) 1.6 (0.4) 2.7 (0.5) 
17 2.9 >4 <1 2.8 4.0 (0.2) 2.1 (0.3) 2.9 (0.1) 1.1 (0.2) 5.4 (0.8) 
18 2.8 2.3 2.9 4.5 (0.5) 1.9 (0.2) 5.3 (1.3) 1.9 (1.2) 3.4 (0.8) 
19 4.1 <1 4.4 4.3 (0.2) 3.4 (0.3) 6.5 (1.2) <1 (0.1) 2.2 (0.3) 
20 7.1 8.1 8.6 6.5 (0.4) <1 (0.0) 8.6 (0.7) 7.2 (1.0) 2.0 (0.1) 
21 4.6 4.1 >4 6.3 6.4 (1.3) 5.4 (2.1) 5.3 (0.3) 6.4 (0.6) 7.3 (1.0) 
22 8.2 >4 7.6 8.6 5.0 (2.4) 6.3 (0.1) 2.1 (0.2) 6.9 (0.1) 4.3 (1.8) 
23 2.7 >4 <1 4.2 1.9 (0.3) 2.8 (0.0) <1 (0.0) 2.7 (0.1) 5.1 (0.2) 
24 5.1 >4 3.3 7.2 >4 4.4 (1.2) 3.7 (1.1) <1 (0.0) 7.3 (0.3) 0.9 (0.0) 
TCRBVPatient 1Patient 2
Lesion 1Lesion 2Lesion 3PBLsLesion 1Lesion 2Lesion 3PBLs
Exp. %ClonesExp. %ClonesExp. %ClonesExp. %Exp. %ClonesExp. %ClonesExp. %ClonesExp. %
3.7 5.7 4.6 >4 4.1 (0.2) 3.5 (1.0) 8.4 (1.3) 8.1 (0.0) 9.2 (1.1) 
4.1 7.9 >4 3.0 3.3 (0.5) 2.7 (0.0) 3 5.4 (1.6) 1 5.0 (0.3) 3 5.9 (1.5) 
6.4 7.4 5.2 >4 7.0 (0.0) 6.6 (1.4) >4 10.6 (0.4) 3.4 (0.3) 3.1 (1.6) 
5.4 >4 9.8 2.3 3.3 (0.7) 6.4 (0.2) 4.0 (0.0) 7.9 (1.1) 5.7 (0.4) 
4.0 9.3 4.5 3.8 (0.1) 9.0 (1.6) >4 7.6 (2.8) 3.9 (0.0) 4.6 (2.2) 
2.8 10.1 6.0 4.2 (1.7) 4.7 (0.1) 4.0 (1.0) 6.9 (0.1) 3.9 (0.2) 
<1 <1 <1 1.5 (0.5) <1 (0.0) <1 (0.1) <1 (0.1) 4.6 (0.6) 
6.7 >4 9.5 2.9 >4 6.2 (0.7) 8.1 (0.5) 9.1 (0.7) 4 9.0 (2.1) 4 3.3 (0.4) 
3.5 2.9 3.1 4.0 (0.3) 2.5 (0.0) 6.3 (2.5) 1.8 (0.1) 4.4 (0.5) 
10 5.6 <1 3.5 4.0 (0.6) 7.3 (0.2) <1 (0.1) 1.4 (0.1) 3.7 (0.4) 
11 <1 <1 3.0 2.9 (0.6) 6.7 (1.4) <1 (0.1) <1 (0.1) 2.6 (0.8) 
12 3.8 <1 3.5 4.8 (0.8) 4.5 (1.6) >4 <1 (0.2) 7.2 (1.7) 3.9 (1.6) 
13 4.2 >4 3.8 2.7 >4 4.3 (1.9) 4.2 (1.2) >4 7.3 (0.2) <4 5.1 (0.7) 3 7.5 (0.7) 
14 6.3 >4 1.1 3.8 3.7 (0.3) 3.9 (1.1) >4 2.5 (0.0) 2 3.6 (1.0) 2 3.4 (0.8) 
15 1.5 <1 1.5 >4 2.5 (0.2) <1 (0.1) 1.5 (0.1) <1 (0.6) 1.5 (0.8) 
16 3.8 4.3 4.6 >4 3.2 (1.5) 2.6 (0.0) 1.0 (0.3) 1.6 (0.4) 2.7 (0.5) 
17 2.9 >4 <1 2.8 4.0 (0.2) 2.1 (0.3) 2.9 (0.1) 1.1 (0.2) 5.4 (0.8) 
18 2.8 2.3 2.9 4.5 (0.5) 1.9 (0.2) 5.3 (1.3) 1.9 (1.2) 3.4 (0.8) 
19 4.1 <1 4.4 4.3 (0.2) 3.4 (0.3) 6.5 (1.2) <1 (0.1) 2.2 (0.3) 
20 7.1 8.1 8.6 6.5 (0.4) <1 (0.0) 8.6 (0.7) 7.2 (1.0) 2.0 (0.1) 
21 4.6 4.1 >4 6.3 6.4 (1.3) 5.4 (2.1) 5.3 (0.3) 6.4 (0.6) 7.3 (1.0) 
22 8.2 >4 7.6 8.6 5.0 (2.4) 6.3 (0.1) 2.1 (0.2) 6.9 (0.1) 4.3 (1.8) 
23 2.7 >4 <1 4.2 1.9 (0.3) 2.8 (0.0) <1 (0.0) 2.7 (0.1) 5.1 (0.2) 
24 5.1 >4 3.3 7.2 >4 4.4 (1.2) 3.7 (1.1) <1 (0.0) 7.3 (0.3) 0.9 (0.0) 
a

Each BV region signal (Exp. %) is expressed as the mean percentage (SD) of the total TCRBV signal. Underlined values represent BV regions expressed at a frequency of 1.8 above the frequency in PBL. Values of <1% are indicated as <1. When more than four clonotypic transcripts were detected by clonotype mapping, this is designated as >4. No clonal transcripts were detected in the PBLs. BV-specific Abs were used to stain the regions shown in bold.

To exclude the possibility that these variations of the TCRBV repertoire were a reflection of differences in the expression of MAA in the analyzed biopsies, RT-PCR was used to analyze for the expression of tyrosinase, MART-1, and gp100 mRNAs. This analysis revealed that all lesions were positive for expression of these transcripts (data not shown). For each of the three lesions, the presence of MART-1 and gp100 proteins was verified by immunohistochemical staining with mAb against these proteins (data not shown).

PBLs and tumor lesions were analyzed for the presence of T cell clonotypes by RT-PCR/DGGE-based TCR clonotype mapping. The detection of clonally expanded T cells via this approach relies on the fact that clonotypic transcripts have no junctional diversity and therefore resolve at a fixed position in the denaturing gradient gel (5). Analysis of the PBLs from both patients did not reveal any distinct bands in the denaturing gradient gel (data not shown; also see Fig. 2). Analysis of the tumor lesions, however, showed the presence of multiple clonotypic TCR transcripts (with numbers ranging from 40 to >60) covering the majority of the BV families 1–24 (Fig. 1 and Table I). Notably, clonotypic transcripts were detected in most BV regions, irrespective of the level of expression. Even regions expressed at a level equaling <1% contained clonotypic TCR transcripts (Table I). Conversely, clonotypic TCR transcripts were absent in some BV regions expressed at high levels, demonstrating that a high level of expression does not necessarily imply the presence of T cell clonotypes (Table I).

FIGURE 2.

TCR clonotype mapping used for a comparative analysis of clonotypes detected in the different lesions from the two patients. Sequence analysis revealed identity for: BV22, lesion 1 (lower band) and lesion 2 (lower band); BV22, lesion 1 (second lower band) and lesion 3 (lower band); BV24, lesions 2 and 3 (second and third lower bands, respectively) from patient 1; BV14, lesion 1 (second lower band) and lesion 3 (lower band) from patient 2.

FIGURE 2.

TCR clonotype mapping used for a comparative analysis of clonotypes detected in the different lesions from the two patients. Sequence analysis revealed identity for: BV22, lesion 1 (lower band) and lesion 2 (lower band); BV22, lesion 1 (second lower band) and lesion 3 (lower band); BV24, lesions 2 and 3 (second and third lower bands, respectively) from patient 1; BV14, lesion 1 (second lower band) and lesion 3 (lower band) from patient 2.

Close modal
FIGURE 1.

TCR clonotype mapping of the T cell infiltrate in an s.c. melanoma metastasis from lesion 3 of patient 1, covering BV families 1–24. PCR products were loaded onto a 20–80% denaturing gradient gel and run for 4.5 h at 160 V at a constant temperature of 58°C. DNA was stained with ethidium bromide and photographed under UV light.

FIGURE 1.

TCR clonotype mapping of the T cell infiltrate in an s.c. melanoma metastasis from lesion 3 of patient 1, covering BV families 1–24. PCR products were loaded onto a 20–80% denaturing gradient gel and run for 4.5 h at 160 V at a constant temperature of 58°C. DNA was stained with ethidium bromide and photographed under UV light.

Close modal

To distinguish whether this clonal expansion of T cells occurs as a systemic immune response or is due to the local activation and outgrowth of T cells, we performed a comparative DGGE analysis of the different melanoma lesions. Surprisingly, as depicted in Fig. 2, most clonotypes were exclusively present in a single lesion. More than fifty clonal TCR transcripts were compared by DGGE; only six transcripts appeared to be accumulated in two different lesions, and none of the clonotypes were present in all three lesions from the same patient.

A number of transcripts resolved at similar positions in the gel, suggesting sequence identity (Fig. 2). These DNA bands were subjected to sequence analysis. Of the six pairs of sequences with identical migration distance, only four were indeed identical. It should be noted that the two different BV4 sequences derived from lesions 1 and 3 of patient 1 shared the same joining sequence and only differed in the junctional region between the V and the J regions.

Staining of the metastatic lesions with mAbs specific for a number of different TCRBV regions was performed to examine the distribution pattern of T cells expressing these individual regions. Representative results are shown in Fig. 3, where lesion 3 of patient 2 was stained with an Ab to either BV13 or BV14. These regions were expressed at 5.1% and 3.6%, respectively, of the total TCR expression, and both contained clonotypic transcripts (Table I). It appears that positive cells were present as small clusters, which is in agreement with the hypothesis that clonotypic expansion occurs in situ (Fig. 3). In contrast, CD3-expressing cells were present throughout the tumor (data not shown).

FIGURE 3.

Clusters of T cells expressing specific TCRBV regions in cutaneous melanoma metastases. The s.c. metastatic lesion 3 of patient 2 was stained with mAbs reacting with either TCRBV13 (A) or TCRBV14 (B). These regions were expressed at 2.7% and 3.8%, respectively, of the total TCR mRNA, and both contained clonotypic transcripts (Table I).

FIGURE 3.

Clusters of T cells expressing specific TCRBV regions in cutaneous melanoma metastases. The s.c. metastatic lesion 3 of patient 2 was stained with mAbs reacting with either TCRBV13 (A) or TCRBV14 (B). These regions were expressed at 2.7% and 3.8%, respectively, of the total TCR mRNA, and both contained clonotypic transcripts (Table I).

Close modal

During the past decade, important insights have been unveiled with regard to the role of T lymphocytes in the immune response to cancer in general and to melanoma in particular (8). The notion that a functional and specific T cell response may be operating in melanoma patients was reinforced by the demonstration of clonotypic T cells in both primary and metastatic melanoma lesions (9, 10). However, only a few attempts have been made to characterize the complexity of the in vivo T cell response. Furthermore, although specific T cell clonotypes have been detected in several lesions from the same patients (11), it remains unknown whether the T cell response to melanoma is maintained by the circulation of specific T cell clones that accumulate at the metastatic site or, alternatively, whether clonotypic expansion is restricted to the individual site.

In the present study, we have attempted to address some of these questions by using the TCR clonotype mapping methodology. Analysis of six melanoma lesions from two patients revealed the presence of multiple clonotypic TCR transcripts in all cases, with numbers of clonotypes ranging from ∼40 forty to >60; furthermore, the majority of the T cell clonotypes were limited to a single lesion only. Thus, the T cell response against malignant melanoma seems to be very complex and to involve a high number of localized T cell clones.

Initial studies have demonstrated the presence of a finite number of T cell clonotypes in lymphocytes infiltrating melanoma (12, 13, 14). These analyses have been conducted using quantitative RT-PCR of TCR V regions, and the subsequent analysis for clonality was limited to overexpressed regions. Recently, methods with considerably higher resolution (e.g., CDR3 size spectra analysis and single-strand conformational polymorphism) have been used to analyze T cell infiltrates in melanoma (15, 16). However, because these methods are rather labor-intensive when used at their full potential, they have often been used to analyze selected TCRBV regions only. The most intensive analysis to date estimated the number of T cell clonotypes in melanoma lesions to be in the range of 2–15 (11). The data presented here suggest that the complexity of the T cell response has been largely underestimated. However, the presence of clonal T cells in most BV regions irrespective of the level of expression suggests that a substantial fraction of in situ T cell clonotypes only proceed through a limited number of cell divisions. A possible explanation could be that specific T cells in situ do not reach the threshold level for full activation due to a low affinity of the TCR (17) or the absence of costimulatory signals. Alternatively, T cell clonotypes detected in regions expressed at low levels could be in the early phase of expansion.

In analogy with our findings, it was recently demonstrated that the numbers of virus-specific T cells during systemic infection were considerably higher than previously anticipated (18, 19, 20). Large clonal expansions of CD8 T cells have been detected in the peripheral blood during acute infection, with viral-specific T cells constituting as much as 40% of the T cells (18, 19, 21). Although a surprisingly high number of T cell clonotypes is present within the lymphocytes infiltrating melanoma, no clonotypes were detected in the blood of the patients, arguing against a major circulation of these clonotypes. This observation may reflect the difference between a localized incident (as in the case of a solid tumor) and a generalized incident (as in the case of a systemic viral infection). Indeed, during a localized viral infection, the specific T cell response is not of similar magnitude as during systemic infections, which is possibly due to the limited Ag load in the blood and lymph nodes (22). The notion of a localized T cell response against melanoma was recently corroborated by the demonstration of a functional dissociation between local and systemic T cell responses during peptide vaccination (23).

The thought that the T cell response against melanoma is localized was further substantiated by immunohistological studies. The melanoma lesions were infiltrated by numerous T cells throughout the tumor, whereas staining with Abs against specific TCRBV regions clearly demonstrated that the T cell clones expressing these regions were located in clusters of cells, suggesting a local clonal expansion. To further investigate whether clonally expanded T cells in situ were local by nature, we compared the clonotypes detected in the different lesions from each patient, with the aim of resolving whether specific T cell clones were identical (i.e., originated from the same clonal expansion). The results demonstrated that the preponderance of the clonotypes were unique (i.e., were present in a single lesion only). The in vivo accumulation of identical T cell clones in different metastases has been reported (11, 16, 24). The present study corroborates these findings, as a minority of the identified T cell clonotypes was present in two different lesions. Whether the capacity of these T cells to migrate to distant metastases is correlated to specialized functions (e.g., regulatory suppressive functions) is not known at the present stage. Furthermore, it is possible that the incidence of detecting identical clonotypes in different lesions would increase if it was possible to follow the infiltrate over time.

Our data indicate that locally expanded T cells execute the T cell response against melanoma, and that the majority of these T cell clonotypes do not recirculate to other metastatic lesions. It is possible that the majority of the clonotypic T cells involved in a cellular response are destined to die by apoptosis, whereas a minority possesses the capability to persist as long-term memory cells. The factors that influence the generation of memory T cells are poorly understood. Nevertheless, an important difference between effector T cells and memory T cells is related to their different growth capacities. The in situ T cell response against melanoma may not support the generation of memory T cells. Thus, the majority of the clonotypic T cells in TILs are effector T cells with a limited potential to proliferate, which implies that the T cells that leave the local environment are cells that have a limited growth capacity. Accordingly, these cells will not expand to a detectable level upon recognition of Ag in distant metastases. This would also explain the repeatedly observed inability to grow in vivo-expanded T cell clonotypes in vitro (25, 26).

The general conclusions from the present study are that the T cell infiltrate of malignant melanoma is exceedingly heterogeneous and composed of a much higher number of different T cell clonotypes than previously appreciated. Furthermore, our data indicate that the vast majority of clonotypic T cells in melanoma TILs, at least in the case of s.c. metastases, may be considered as being strictly localized.

We thank Prof. A. Svejgaard and B. K. Jakobsen (Department of Clinical Immunology, University Hospital, Copenhagen, Denmark) for HLA-typing and D. Lützhøft for preparation of samples for the isolation of RNA.

1

We thank the Danish Melanoma Group for continued interest and support. This work was supported by grants from the Danish Cancer Society.

3

Abbreviations used in this paper: MAA, melanoma-associated Ags; TIL, tumor-infiltrating lymphocyte; DGGE, denaturing gradient gel electrophoresis; BV, β-variable; gp, glycoprotein.

1
Van den Eynde, B. J., T. Boon.
1997
. Tumor antigens recognized by T lymphocytes.
Int. J. Clin. Lab. Res.
27
:
81
2
thor Straten, P., J. C. Becker, T. Seremet, E. B. Bröcker, J. Zeuthen.
1996
. Clonal T-cell responses in tumor-infiltrating lymphocytes from both regressive and progressive regions of primary human melanoma.
J. Clin. Invest.
98
:
279
3
Sensi, M., C. Farina, C. Maccalli, R. Lupetti, G. Nicolini, A. Anichini, G. Parmiani, D. Berd.
1997
. Clonal expansion of T lymphocytes in human melanoma metastases after treatment with a hapten-modified autologous tumor vaccine.
J. Clin. Invest.
99
:
710
4
thor Straten, P., E. Ralfkiaer, J. Hendriks, T. Seremet, G. L. Vejlsgaard, J. Zeuthen.
1998
. T-cell receptor variable region genes in cutaneous T-cell lymphomas.
Br. J. Dermatol.
138
:
3
5
thor Straten, P., A. Barfoed, T. Seremet, I. Säterdal, J. Zeuthen, P. Guldberg.
1998
. Detection and characterization of α/β clonality by denaturing gradient gel electrophoresis (DGGE).
Biotechniques
25
:
244
6
Johnston, R. F., S. C. Pickett, D. L. Barker.
1990
. Autoradiography using storage phosphor technology.
Electrophoresis
11
:
355
7
thor Straten, P., A. F. Kirkin, T. Seremet, J. Zeuthen.
1997
. Expression of transporter associated with antigen processing 1 and 2 (TAP1/2) in malignant melanoma cell lines.
Int. J. Cancer
70
:
582
8
Boon, T., P. van der Bruggen.
1996
. Human tumor antigens recognized by T lymphocytes.
J. Exp. Med.
183
:
725
9
Kawakami, Y., S. Eliyahu, C. Jennings, K. Sakaguchi, X. Kang, S. Southwood, P. F. Robbins, A. Sette, E. Appella, S. A. Rosenberg.
1995
. Recognition of multiple epitopes in the human melanoma antigen gp100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression.
J. Immunol.
154
:
3961
10
Salvi, S., F. Segalla, S. Rao, F. Arienti, M. Sartori, G. Bratina, E. Caronni, A. Anichini, C. Clemente, G. Parmiani, M. Sensi.
1995
. Overexpression of the T-cell receptor β-chain variable region TCRBV14 in HLA-A2-matched primary human melanomas.
Cancer Res.
55
:
3374
11
Puisieux, I., J. Even, C. Pannetier, F. Jotereau, M. Favrot, P. Kourilsky.
1994
. Oligoclonality of tumor-infiltrating lymphocytes from human melanomas.
J. Immunol.
153
:
2807
12
thor Straten, P., J. Schøller, K. Hou Jensen, J. Zeuthen.
1994
. Preferential usage of T-cell receptor αβ-variable regions among tumor-infiltrating lymphocytes in primary human malignant melanomas.
Int. J. Cancer
56
:
78
13
Mackensen, A., L. Ferradini, G. Carcelain, F. Triebel, F. Faure, S. Viel, T. Hercend.
1993
. Evidence for in situ amplification of cytotoxic T-lymphocytes with antitumor activity in a human regressive melanoma.
Cancer Res.
53
:
3569
14
Sensi, M., C. Farina, C. Maccalli, A. Anichini, D. Berd, G. Parmiani.
1998
. Intralesional selection of T cell clonotypes in the immune response to melanoma antigens occurring during vaccination.
J. Immunother.
21
:
198
15
Pannetier, C., J. Even, P. Kourilsky.
1995
. T-cell repertoire diversity and clonal expansions in normal and clinical samples.
Immunol. Today
16
:
176
16
Hishii, M., D. Andrews, L. A. Boyle, J. T. Wong, F. Pandolfi, P. J. van den Elsen, J. T. Kurnick.
1997
. In vivo accumulation of the same anti-melanoma T cell clone in two different metastatic sites.
Proc. Natl. Acad. Sci. USA
94
:
1378
17
Gervois, N., Y. Guilloux, E. Deiz, F. Jotereau.
1996
. Suboptimal activation of melanoma infiltrating lymphocytes (TIL) due to low avidity of TCR/MHC-tumor peptide interaction.
J. Exp. Med.
183
:
2403
18
Callan, M. F., N. Steven, P. Krausa, J. D. Wilson, P. A. Moss, G. M. Gillespie, J. I. Bell, A. B. Rickinson, A. J. McMichael.
1996
. Large clonal expansions of CD8+ T cells in acute infectious mononucleosis.
Nat. Med.
2
:
906
19
Altman, J. D., P. A. Moss, P. J. R. Goulder, D. H. Barouch, M. G. McHeyzer-Williams, J. I. Bell, A. J. McMichael, M. M. Davis.
1996
. Phenotypic analysis of antigen-specific T lymphocytes.
Science
274
:
94
20
Callan, M. F., L. Tan, N. Annels, G. S. Ogg, J. D. Wilson, C. A. O’Callaghan, N. Steven, A. J. McMichael, A. B. Rickinson.
1998
. Direct visualization of antigen-specific CD8+ T cells during the primary immune response to Epstein-Barr virus in vivo.
J. Exp. Med.
187
:
1395
21
Pantaleo, G., J. F. Demarest, H. Soudeyns, C. Graziosi, F. Denis, J. W. Adelsberger, P. Borrow, M. S. Saag, G. M. Shaw, R. P. Sekaly, et al
1994
. Major expansion of CD8+ T cells with a predominant Vβ usage during the primary immune response to HIV.
Nature
370
:
463
22
Flynn, K. J., G. T. Belz, J. D. Altman, R. Ahmed, D. L. Woodland, P. C. Doherty.
1998
. Virus-specific CD8+ T cells in primary and secondary influenza pneumonia.
Immunity
8
:
683
23
Lee, K. H., M. C. Panelli, C. J. Kim, A. I. Riker, M. P. Bettinotti, M. M. Roden, P. Fetsch, A. Abati, S. A. Rosenberg, F. M. Marincola.
1998
. Functional dissociation between local and systemic immune response during anti-melanoma peptide vaccination.
J. Immunol.
161
:
4183
24
Strohal, R., C. Brna, U. Mossbacher, G. Fischer, H. Pehamberger, G. Stingl.
1998
. First comparative delineation of the T cell receptor repertoire in primary and multiple subsequent/coexisting metastatic melanoma sites.
J. Invest. Dermatol.
111
:
1085
25
Dietrich, P. Y., P. R. Walker, V. Schnuriger, P. Saas, G. Perrin, M. Guillard, C. Gaudin, A. Caignard.
1997
. TCR analysis reveals significant repertoire selection during in vitro lymphocyte culture.
Int. Immunol.
9
:
1073
26
Faure, F., J. Even, P. Kourilsky.
1998
. Tumor-specific immune response: current in vitro analyses may not reflect the in vivo immune status.
Crit. Rev. Immunol.
18
:
77