Abstract
Fibrosis, microvascular fibroproliferative alterations, and autoantibody production are the main features of systemic sclerosis (SSc), and all of them can be explained by cytokine production by activated T cells. However, little is known about the role of T cells in the pathogenesis of SSc, and there is no information on the Ag(s) that elicits such activation. To determine whether T cells infiltrating the skin biopsies of patients with SSc are oligoclonal, β-chain TCR transcripts from T cells infiltrating the skin of five patients with SSc of recent onset were amplified by either Vβ-specific PCR or nonpalindromic adaptor PCR. The resulting PCR products were subsequently cloned and sequenced. High proportions of identical β-chain TCR transcripts ranging from 43 to 90% of those sequenced were found in five patients, strongly suggesting the presence of oligoclonal T cells in these infiltrates. A dominant T cell clone was found to be clonally expanded in skin biopsies obtained from a single patient with SSc at three different times (0, 8, and 13 mo earlier) and from three different skin regions. β-chain TCR transcripts from PBMC from normal donors (methodological control) were unique when compared with each other, typical for polyclonal populations of T cells. The finding of oligoclonal T cells infiltrating the skin of patients with SSc suggests that these T cells have undergone proliferation in situ in the skin and clonal expansion in response to as yet unidentified Ag(s). These results suggest that T cells are involved in the pathogenesis of the disease.
Systemic sclerosis (SSc)5 is characterized by widespread fibrosis, microvascular fibroproliferative alterations, and autoantibody production; however, the pathogenic mechanisms leading to these changes are largely unknown (1, 2, 3). Although there is evidence of activation of fibroblasts, monocytes, endothelial cells, eosinophils, and B cells in SSc, a common event of all three characteristic features of the disease appears to be the involvement of T cells. There is a cellular infiltration of the skin, consisting of T cells, macrophages, and mast cells, early in the course of the disease (4, 5, 6, 7). T cells are also found in lung tissues from SSc patients with lung involvement (8). Peripheral blood T cells in SSc exhibit signs of activation and express activation Ags, such as IL-2R (CD25), HLA-DR, and CD29 (9, 10), and increased levels of protooncogenes (11). Increased levels of soluble IL-2R were found in suction blister fluid in the skin of patients with SSc (12). All these findings suggest in vivo activation of T cells. T cells are necessary for the production of anti-topoisomerase I autoantibodies (formerly Scl 70) (13), which are characteristic of SSc. However, the production of these autoantibodies may be autoimmune.
It appears that T cells infiltrating SSc lesions have undergone activation and proliferation, perhaps in response to unknown Ag(s). Although there is no information on these Ag(s), fetal cells of various hemopoietic cell lineages have been identified in the peripheral blood and the skin of women with SSc who had been previously pregnant, as a result of microchimerism (14, 15, 16, 17, 18, 19, 20). Microchimerism of HLA-disparate maternal cells can persist in both SSc patients and normal donors (21). Sources of engraftment of cells in both men and women may include the fetus, the mother, a twin sibling, or a blood transfusion. Activation of these fetal T cells in the mother, in response to an unknown antigenic stimulus, or because of breaking of the tolerance mechanisms, which regulate a steady-state equilibrium that permits the microchimerism to exist, may result in the induction of a chronic fetal antimaternal graft-vs-host disease (GVHD), which is manifested as SSc (14, 15, 16, 17, 18, 19, 20). Microchimerism of maternal cells has been identified in women who have not been previously pregnant or in men, and it may persist in both SSc patients and healthy subjects (20, 21, 22, 23). Similarly, activation of these maternal T cells in the offspring, by the mechanisms described above, may also result in the induction of a chronic maternal anti-offspring GVHD (20, 21, 22, 23). An allospecific T cell response of fetal T cells to maternal alloantigen(s) (or of maternal T cells to offspring alloantigens) may be responsible for the appearance of SSc. In this event, the alloantigen could be a putative SSc Ag. Strong similarities between GVHD and SSc have been identified, including fibrosis, microvascular fibrointimal proliferation, and autoantibody (anti-topoisomerase I) production (24).
Activated T cells produce Th2 cytokines, which appear to cause fibrosis in SSc. IL-4, which is increased in the peripheral blood of patients with SSc (25, 26), induces production of collagen and other extracellular matrix macromolecules by fibroblasts in vitro (27, 28, 29). We have demonstrated the presence of increased levels of alternatively spliced IL-4 (IL4δ2) transcripts in PBMC from patients with SSc (26). Overexpression of IL-4 in transgenic mice under the control of the insulin promoter in pancreatic Langerhans cells results in local fibrosis (30). Anti-IL-4 Ab prevents GVHD in mice (31) and reduces hepatic fibrosis in Schistosoma-infected mice (32). Finally, IL-4 induces the production of TGF-β (33), which causes tissue fibrosis and fibrointimal proliferation of blood vessels (34) and could explain the tissue fibrosis and vascular injury of SSc. Although the role of T cells in the widespread fibrosis, the microvascular fibroproliferative alterations, and the autoantibody production in SSc are well documented and can be explained by the cytokines produced by these activated T cells, there is no information on the Ag(s) that induce T cell activation in SSc.
T cells can be activated through their TCR, which recognizes antigenic peptides in association with HLA on APCs (35). In SSc it is not known whether the activation of T cells is Ag induced. However, SSc has been found to be associated with HLA (36, 37, 38), which supports the concept of an Ag-driven T cell expansion. To examine this hypothesis, we sequenced β-chain TCR transcripts from skin biopsies of patients with SSc of recent onset. The presence of substantial proportions of identical β-chain TCR transcripts in T cells infiltrating affected skin of patients with SSc strongly suggests the presence of an Ag-driven proliferation and clonal expansion of T cells.
Materials and Methods
Patients
Skin biopsies from five patients with SSc of recent onset (<18 mo from the appearance of clinically detectable skin induration) who were followed up at the Scleroderma Center of Thomas Jefferson University Hospital (Philadelphia, PA) were used in this study. The removal of these skin biopsies from patients with SSc and their use in in vitro biological studies were approved by the Institutional Review Board of Thomas Jefferson University Hospital. All patients and normal controls provided informed consent. These five patients were designated S70, S94, S162, S168, and S169. They all fulfilled the criteria for classification of SSc described by the Scleroderma Subcommittee of the American College of Rheumatology (39). From one patient (S169), three different SSc skin biopsy specimens were available. One (S169A) of these skin biopsy specimens was removed 13 mo earlier than the S169C skin biopsy specimen, and the second (S169B) was removed 8 mo earlier than the S169C skin biopsy. The demographics of these patients are shown in Table I⇓. Skin biopsies were obtained from the leading edge of affected skin in the forearms of these SSc patients. The specimens were snap-frozen in liquid nitrogen and were stored in liquid nitrogen until used. Skin biopsies were usually divided in two generally equal fractions. One was used for histology and immunofluorescence studies and the other for preparation of RNA. Peripheral blood (30 ml) from normal donors was collected in heparinized tubes and PBMC were isolated by centrifugation on Ficoll-Hypaque (Amersham Pharmacia Biotech, Piscataway, NJ), and kept at −80°C until used. These PBMC were used as methodological controls.
Demographics of patients with SSca
Immunofluorescence
Six-micron-thick cryostat sections of skin biopsies from two patients (S169A and S70) were stained for CD3 using the indirect immunofluorescence method and anti-CD3 mAb (clone UCHT-1; DAKO, Carpinteria, CA). Briefly, sections were incubated first with the anti-CD3 mAb (1/100 dilution) for 40 min, washed, and then incubated with Cy2-conjugated goat anti-mouse Ab (1/50 dilution; Jackson ImmunoResearch Laboratories, West Grove, PA) for 40 min. Sections were mounted and examined under an epifluorescent microscope. Results were expressed as average number of positive cells per high-power field (hpf; ×400).
Synthesis of cDNA Vβ-specific PCR
Total RNA was prepared using RNAzol B according to supplier’s instructions (Tel-Test, Friendswood, TX). cDNA from three patients (S162, S168, and S169) was synthesized from 3 μg of RNA using Superscript II reverse transcriptase and oligo(dT) as a primer (Life Technologies, Rockville, MD) at 42°C for 50 min in a 20-μl reaction (26). The mixtures were then heated at 70°C for 15 min to inactivate reverse transcriptase and incubated with RNase H at 37°C for 20 min to remove RNA. Finally, cDNA was diluted 1/2 and kept at −30°C until used. The presence of T cells in skin biopsies was determined by PCR in a 480 ThermoCycler (PerkinElmer, Wellesley, MA) as described (24). β-Chain TCR cDNA was amplified using Vβ-specific primers (Table II⇓) (40) and a nested PCR to diminish cross-reactivity with other members of the Ig supergene family (41). The first amplification of cDNA (50 ng of RNA equivalents) was conducted in a reaction containing 50 mM Tris-HCl (pH 8), 100 mM NaCl, 0.1 mM EDTA, 1 mM DTT, 1.5 mM MgCl2, 50% glycerol, 1% Triton X-100, and 2.5 U Taq DNA polymerase (Promega, Madison, WI). A Vβ-specific oligonucleotide was used as the 5′ amplification primer and the Cβ2 oligonucleotide as the 3′ primer (Table II⇓). The reaction was conducted in 25 cycles with each cycle at 94°C for 45 s, 60°C (65°C for β-actin) for 45 s, and 72°C for 90 s, and a final extension of 7 min at 72°C. Two microliters of the PCR product was used in the second amplification conducted under the same conditions, with the exception that the Cβ1 primer was used as the 3′ amplification primer (Table II⇓). PCR products were visualized on ethidium bromide-stained 1.6% agarose gels after electrophoresis.
Vβ-specific amplification primers and NPA-PCR adapters
NPA-PCR
Total RNA from four specimens (S70, S94, S169A, and S169B) was isolated using a RNA isolation kit (Stratagene, La Jolla, CA) according to the supplier’s instructions. Double-stranded cDNA was synthesized from oligo(dT)-NotI (Promega)-primed total RNA, using the Superscript II cDNA synthesis kit (Life Technologies). The double-stranded cDNA was blunt-ended using T4 DNA polymerase in the last step of cDNA synthesis for efficient adapter ligation. Nonpalindromic adaptor (NPA)-PCR was performed as previously described (42, 43, 44, 45, 46), with minor modifications. Double-stranded blunt-ended cDNA was ligated by incubating with T4 DNA ligase at 16°C for 14 h, with the NPA (two complimentary oligonucleotides: EcoRI-XmnI strand and XmnI G strand (phosphorylated) (Table II⇑). The two strands of the adaptor were preannealed to each other. The adaptor was ligated at both the 5′ end and the 3′ end of the dsDNA. The ligated adaptor was removed from the 3′ end of cDNA by digestion with the NotI restriction endonuclease (20 U) for 2 h at 37°C. The product was purified by centrifugation on a G-50 column (5 Prime→3 Prime, Boulder, CO). A nested PCR design was used for the amplification of the NotI-digested cDNA to eliminate cross-reactivity with other members of the Ig supergene family (41). It is unlikely that this cross-reactivity will occur in several parts of the molecule. Two amplification rounds were conducted as described (42, 43, 44, 45, 46). In both amplification rounds of the nested PCR, the adapter primer EcoRI-XmnI was used as the 5′ primer. The Cβ3 primer was used as the 3′ primer for the first round of amplification, and the Cβ2 primer was used for the second round of amplification. The Cβ2 primer is located 5′ to the Cβ3 primer. The NPA-PCR was carried out in a total volume of 100 μl. Both rounds of amplification were conducted in 30 cycles. Each cycle included the following steps: denaturation (94°C for 1 min), annealing (60°C for 1 min), and elongation (72°C for 1 min), followed by a 10-min final incubation at 72°C. The product of the first amplification was purified by PCR Select III Spin Column (5 Prime→3 Prime). cDNA from the second amplification was separated by 0.04 M Tris acetate, 0.001 M EDTA (TAE) gel electrophoresis and purified with NaI (Geneclean kit; Bio 101, Vista, CA).
Cloning and sequencing of PCR products
β-Chain TCR transcripts were sequenced as described (26). Briefly, amplified β-chain TCR transcripts were separated from the TAE/low-melting point agarose gel, purified using NaI (Geneclean kit), and ligated into the pCR2.1 vector with overhanging single 3′ deoxythymidine residues (Invitrogen, Carlsbad, CA). This vector was then used to transform INVaF cells (Invitrogen). Briefly, INVaF Escherichia coli cells were incubated with the vector on ice for 45 min and then submitted to 42°C heat shock for 45 s, followed by resting on ice for 2 min. Next, cells were incubated in 450 μl SOC medium (20 g bacto-tryptone, 5 g bacto-yeast extract, 0.5 g NaCl, 2.5 M KCl, 0.1 M MgCl2, and 20 mM glucose) at 37°C for 1 h and plated on agar plates containing X-gal (50 μg/ml) and ampicillin (100 μg/ml). After overnight culture, white microbial colonies were cultured in Terrific broth (Fisher Scientific, Atlanta, GA) and mini plasmid preparations were prepared using the alkaline lysis method and Wizard DNA purification system (Promega). Plasmids bearing the TCR Vβ inserts were identified by PCR and agarose gel electrophoresis. β-chain TCR transcripts were subjected to ThermoSequenase dye terminator sequencing PCR (Amersham, Cleveland, OH), and the PCR products were purified using Centricep spin columns (Princeton Separations, Adelphia, NJ) and analyzed by 6% PAGE using an automated 373A DNA sequencer (PE Applied Biosystems, Foster City, CA).
The maximum number of unique β-chain TCR transcripts in humans is ∼1012 (47). Therefore, the probability to find by chance two identical copies of β-chain TCR transcripts in a given independent sample of T cells is negligible. However, during transformation of DH5α-E. coli-competent cells, the plasmid/cell mixture was subjected to heat shock treatment at 42°C for 45 s, followed by incubation on ice for 2 min and then growth for 1 h in SOC medium at 37°C before plating the colonies. Under the log phase growth of E. coli (ideal growth conditions) it can undergo a division in 20 min, which could result in two doublings within 60 min (48). However, because of the heat shock, E. coli cells do not immediately enter the log phase, although the unlikely possibility for a few E. coli-transformed cells to double before plating does exist. Therefore, a doublet, i.e., identical TCR sequences from two different colonies, may be the result of a single transfected E. coli which doubled before plating or may reveal a clonal expansion. It appears that possible doubling of singly transfected E. coli cells before plating is rather infrequent. We have amplified by NPA-PCR or by Vβ-specific PCR, cloned, and sequenced over 150 β-chain TCR transcripts from PBMC from normal donors (46). All β-chain TCR transcripts were unique when compared with each other, with the exception of one clone, which appeared in duplicate (46).
Computer analysis and comparison of sequences
V, D, J, and C regions found in β-chain TCR transcripts obtained from patients with SSc or normal donors were identified and compared with GenBank and EMBL databases using the basic local alignment search tool sequence alignment software (National Center for Biotechnology Information, National Institutes of Health, Bethesda, MD) (49).
Results
Histological evaluation of H&E-stained slides revealed that the skin biopsy specimens from all five patients with SSc had mononuclear infiltrates. Amplification of β-chain TCR transcripts from specimens S162, S168, and S169 by Vβ-specific RT-PCR, using Vβ-specific 5′ primers to amplify each one of 24 Vβ families (Table II⇑), demonstrated very restricted usage of Vβ in all three specimens. Visualization of the PCR products from the 24 Vβ-specific PCRs, on ethidium bromide-stained 1.6% agarose gels after electrophoresis, revealed that only the Vβ13, Vβ14, and Vβ21 gene segments were detected in skin biopsy specimen from SSc patient S162, only the Vβ13 segment was detected in skin biopsy specimen from SSc patient S168, and only the Vβ21 segment was detected in skin biopsy specimen from SSc patient S169 (data not shown). Sequence analysis of these β-chain TCR transcripts after cloning showed large proportions of identical β-chain TCR transcripts in all three patients, which ranged from 42.8 to 89% (Table III⇓), demonstrating the presence in all three patients of strong clonally expanded populations of T cells.
Multiple identical β-chain TCR transcripts expressed in skin biopsies from five patients with SSc
In particular, Vβ13-specific PCR amplified various members of the Vβ13 family (Vβ13.1, Vβ13.2, and Vβ13.9) from skin biopsy specimen from patient S162. Clone 13.1-2 (Vβ13.9 Dβ2 Jβ2.2; CASSIRHEFWTSGSYPGELF) accounted for 9 of 21 (42.8%) of the Vβ13 clones sequenced. An additional clone, 13.2-2 (Vβ13.2 Dβ2 Jβ2.2; CASSIRHEFWTSGSYPGELF), which was present in three copies (3 of 21; 14%) used an identical CDR3 sequence to clone 13.1-2 but a different Vβ13 allele (Vβ13.9 vs Vβ13.2; Table III⇑). The same CDR3 was used in 12 of 21 (57%) clones. Two other Vβ13 clones were also clonally expanded. Clone 13.3-2 (Vβ13.1 Dβ1 Jβ2.3; CASSYVLAGTSLPSTLETQZYFG) accounted for 5 of 21 (24%), and clone 13.4-2 (Vβ13.1 Dβ1 Jβ1.5; CASSYVLAGTSLPSTLETQZYFG) accounted for 3 of 21 (14%) of the Vβ13 clones sequenced.
Clone 14.1-2 (Vβ14.1 Dβ1 Jβ1.5; CASSPSVSNQPQ) accounted for 20 of 23 (87%) Vβ14 TCR transcripts sequenced from the skin biopsy specimen from patient S162, demonstrating a very strong clonal expansion (Table III⇑). Two TCR Vβ21 clones (21.1-2 (Vβ21.2 Dβ1 Dβ2 Jβ2.1; CASSLALGRELGEQ) and 21.2-2 (Vβ21.2 Dβ1 Dβ2 Jβ2.1; CASSLALGRELGEQ)) were also clonally expanded, and comprised, respectively, 10 of 20 (50%) and 9 of 20 (45%) of the Vβ21 clones sequenced from SSc patient S162. These two clones were identical with the exception of one base in the J region (A to T). A third Vβ21 clone, 21.3-2 (Vβ21.2 Dβ1 Dβ2 Jβ2.1; CASSLALGREPGEQ), was present only in one copy, and there was only one nucleotide difference in the J region between this clone and clone 21.2-2 (T to C) (Table III⇑). There was only one amino acid difference (in J region; P to L; Table III⇑) between clone 21.3-2 and clones 21.1-2 and 21.2-2.
In skin biopsy from patient S168 two strong clonally expanded TCR Vβ13 clones were found, one of which accounted for 11 of 20 (55%) Vβ13 clones and the other for 9 of 20 (45%) Vβ13 clones (Table III⇑). These two clones used different Vβ gene segments (Vβ13.1 and Vβ13.6) but shared the same CDR3 (CASSYLLGGNYGYT), which in this patient, S168, was identical in all 20 of 20 (100%) TCR Vβ13 clones sequenced.
In skin biopsy from patient S169, clone 21.1-9 was clonally expanded and accounted for 17 of 19 (89%) Vβ21 clones sequenced (Table III⇑). In this patient all TCR Vβ21 clones shared the LGE motif. Clone 21.1-9 (CDR3: CASSLALGRELGEQFF), which was clonally expanded, was different by only one base (A to G) from clone 21.2-9 (CDR3: CASSLALGRGLGEQFF), resulting in a single amino acid change (E to G). Similarly, clone 21.1-9 (CDR3: CASSLALGRELGEQFF) was different only by one base (T to C) from clone 21.3-9 (CDR3: CASSLAPGRELGEQFF), resulting in a single amino acid change (L to P). Clonally expanded TCR transcripts from patients S162 and S169 used the same CDR3 but different Vβ21 segments (Vβ21.2 vs Vβ21.1).
Sequence analysis of β-chain TCR transcripts from skin biopsies from patient S70 after NPA-PCR amplification, cloning, and sequencing revealed a very strong clonal expansion (Table III⇑). Clone 70.1 (Vβ14.1 Dβ2.1 Jβ2.5; CDR3: CASSLTPRTDQGTQYF) accounted for 19 of 20 (95%) β-chain transcripts sequenced.
Sequence analysis of β-chain TCR transcripts from skin biopsies from patient S94 after NPA-PCR amplification, cloning, and sequencing also revealed a very strong clonal expansion. Seventeen of 21 (81%) transcripts were identical (clone 94.1, Vβ21.1 Dβ1.1 Jβ1.6; CDR3: CASSFGQGVSPLH).
From patient S169 two additional SSc skin biopsies to the S169C specimen were available. One (S169A) of these skin biopsy specimens was removed 13 mo earlier than the S169C skin biopsy, and the other (S169B) was removed 8 mo earlier than the S169C skin biopsy. A clonally expanded β-chain TCR transcript (clone 21.1-9) was already identified in skin biopsy S169C, after Vβ21-specific PCR amplification, followed by cloning and sequencing (Table III⇑). This clone accounted for 17 of 19 (89%) transcripts sequenced (Tables III⇑ and IV⇓). β-Chain TCR transcripts from the S169A skin biopsy, which was obtained 13 mo earlier than the S169A biopsy, and from the S169B skin biopsy, which was obtained 8 mo earlier than the S169A biopsy, were amplified by NPA-PCR, a different method from that used with the S169C biopsy, and the amplified transcripts were cloned and sequenced. The same clonally expanded β-chain TCR transcript (clone 21.1-9) that was already identified in skin biopsy S169C was also clonally expanded in the S169A biopsy, where it accounted for 19 of 20 (95%) of the β-chain TCR transcripts sequenced, and in the S169B biopsy, where it accounted for 9 of 23 (39%) of the β-chain TCR transcripts sequenced (Table IV⇓). Amplification, cloning, and sequencing of β-chain TCR transcripts from skin biopsies S169A and S169B were conducted 8 mo later than that of those from skin biopsy S169C (all specimens were archived), making contamination very unlikely.
Identical β-chain TCR transcripts were expressed in skin biopsies from one patient (S169) with SSc at different time pointsa
Twenty Vβ13 TCR transcripts, 18 Vβ14 TCR transcripts, and 20 Vβ21 TCR transcripts were obtained after amplification by Vβ-specific PCR, followed by cloning and sequencing of β-chain TCR transcripts from PBMC from healthy donors (Table V⇓). All sequences were unique when compared with each other, as expected with TCR transcripts derived from polyclonal populations of T lymphocytes (42, 43, 44, 45, 46). Furthermore, we have sequenced >200 β-chain TCR transcripts from PBMC from normal donors after amplification by NPA-PCR, followed by cloning and sequencing (Refs. 42, 43 , and 46 and unpublished results). These results have been shown elsewhere and they will not be repeated in this work. All of these PBMC sequences were unique when compared with each other, as expected with TCR transcripts derived from polyclonal populations of T cells (Refs. 42, 43 , and 46 and unpublished results).
Vβ13, Vβ14, and Vβ21 transcripts amplified by Vβ-specific PCR from PBMC from a normal donor are unique when compared to each other, and typical of polyclonal populations of T cells
Continued
It could be argued that if amplification by two PCR cycles is conducted from very few T cells then it is possible that each pair of primers will amplify signals from only few T cells, providing results that may resemble those in Table III⇑. We have demonstrated that this is not the case and that our results represent true clonal expansions of T cells. First, each skin biopsy from patients with SSc was divided into two fractions; one was used for histology and immunofluorescence and the other for RNA preparation. RNA was prepared with a yield of ∼3 μg per specimen, which represents ∼3 × 106 cells. From this RNA we used 50 ng, which represents ∼5 × 104 cells, for PCR amplification. It should be emphasized that the representation of the TCR clonotypes does not change between the sample of 3 μg of RNA and the sample of 50 ng that we used for PCR amplification, cloning, and sequencing. The TCR clonotypes, particularly the expanded ones, present in 3 μg of RNA are also present in the 50 ng of RNA. The ratio of the various TCR clonotypes to the other clonotypes present does not change. What does change is the absolute numbers of the TCR copies present. Second, we determined the number of T cells present in skin biopsy fractions used for RNA preparations from two patients with SSc (S169A and S70) by immunostaining using an anti-CD3 mAb. The average number of CD3+ T cells in biopsy specimen S169A was 5.3 per hpf (×400) or 27 T cells/mm2 (the radius of ×400 hpf is 0.5 mm). Given that the section area was 3 × 10 mm2, the total number of CD3+ T cells is 810 per section. The average number of CD3+ T cells in biopsy specimen S70 was 4.3 per hpf (22 T cells/mm2) or ∼660 CD3+ T cells per section (the section area was 3 × 10 mm2). Given the thickness of skin biopsy specimens of ∼3 mm and that the cryostat sections of skin biopsies used for the immunofluorescence determinations were 6 μm thick, the total number of CD3+ T cells used for RNA isolation in S169A and S70 biopsy specimens is estimated to be 4.05 × 105 and 3.3 × 105, respectively. Because 3 μg of RNA was recovered per specimen and this represents ∼3 × 106 cells, CD3+ T cells accounted in these biopsies for 13.5 and 11%, respectively (mean 12.25%), of the total cells used for RNA isolation. Fifty nanograms of RNA, which represents ∼5 × 104 cells, were used for PCR amplification. Approximately 12% of these cells, i.e., 6125 cells, are CD3+ T lymphocytes in these SSc biopsies.
Control experiments were conducted to address the question that amplification by two PCR cycles, when conducted from very few numbers of T cells, may permit each pair of primers to amplify transcripts from very few T cells. If this had been the case, it is possible that erroneous clonal expansions would have been obtained. Cell mixtures (a total of 1 × 106 cells) comprised of various proportions of the ovarian tumor cell line CAOV3 (which does not express TCR transcripts) and peripheral blood T cells from a normal donor were prepared containing, respectively, 6 and 0.6% T lymphocytes. RNA was prepared from the mixture as described in Materials and Methods and 50 ng of RNA from each mixture, containing 12,000 and 1,200 T cells, respectively, was used for amplification by NPA-PCR, cloning, and sequencing. Sequence analysis revealed the presence of unique transcripts when compared with each other, typical of polyclonal populations of T cells (data not shown). These cells (1,200 T lymphocytes) were five times lower than those present (6,125 T lymphocytes) in SSc skin biopsies.
Similar results were obtained after Vβ-specific amplification, followed by cloning and sequencing. RNA was prepared from mixtures (each comprised of a total of 1 × 106 cells) of the ovarian tumor cell line CAOV3 and peripheral blood T cells from a normal donor containing, respectively, 6, 2.4, and 0.15% T lymphocytes. A total of 50 ng of RNA from each mixture, containing 12,000, 4,800, and 300 T cells, respectively, were used for Vβ2-specific amplification, which was selected as an example of a Vβ family, followed by cloning and sequencing. Sequence analysis revealed the presence of unique Vβ2 TCR transcripts when compared with each other, typical of polyclonal populations of T cells (data not shown) from the mixtures that contained 12,000 and 4,800 T cells. These T cell numbers (4,800) were lower than those present (6,125 cells) in SSc skin biopsies. Vβ2+ T cells represent ∼8% of the Vβ+ T cells (50). Therefore, 384 Vβ2+ T cells were present in the mixture of CAOV3 and peripheral blood T cells from the normal donor and 490 Vβ2+ T cells were present in SSc skin biopsies. From the mixture, which contained only 300 T cells or 24 Vβ2+ T cells, a more restricted pattern was observed, consisting of a Vβ2+ transcript that appeared in three copies, three other Vβ2+ TCR transcripts that appeared in duplicate, and six other Vβ2+ TCR transcripts that appeared in single copies. These results do not demonstrate an oligoclonal expansion, for the reasons described in Materials and Methods (46), even when such a low number, 24, of Vβ2+ T cells was amplified. These results confirmed that the oligoclonality of T cells detected in skin biopsies from patients with SSc was due to real clonal expansions and not to amplifications of TCR transcripts from just a few T cells.
In this study we identified strong highly selective clonal expansions of β-chain TCR transcripts in skin biopsy specimens from patients with SSc. Three conserved amino acid motifs within the CDR3 region of the β-chain TCR within individual patients and between different patients have been identified. These are the LG motif, which was present in the CDR3 of patients S162 (16%), S168 (100%), and S169 (79%); the LAL motif, which was present in the CDR3 of patients S162 (16%) and S169 (74%); and the QG motif, which was present in the CDR3 of patients S70 (95%) and S94 (81%).
Comparison of all sequences obtained to those in the GenBank/EMBL database using the National Center for Biotechnology Information basic local alignment search tool software revealed that all sequences obtained in this study are novel. However, the clonally expanded clone 14.1-2 from patient S162 shared Jβ sequences with T cell clones isolated from the peripheral blood and the bronchoalveolar lavage of a patient with beryllium-induced lung disease (GenBank GI no. 5882111; provided by A. P. Fontenot, M. T. Falta, L. S. Newman, and B. L. Kotzin, National Jewish Medical and Research Center, Denver, CO). In addition, all TCR Vβ13 clones from patient S168 shared the GG amino acid motif of the N-D-N region and the Jβ amino acid sequences (GGNYGYT) with the CDR3 of a nucleosomal DNA-induced T cell line (51) and a T cell clone from a 2-day-old infant with HIV-1 infection (52).
The clonally expanded β-chain TCR transcript clone 70.1 shared substantial homology (CASSLTP) with a T cell clone isolated from the T cell infiltrate of active multiple sclerosis lesions (53). Clone 70.2 (CDR3: CASSQDGEDMNTEA) shared substantial homology with an HTLV-1-specific T cell clonotype (CDR3: CASSQEKDMNTEA) (GenBank GI no. 11527700; provided by M. Saito, Imperial College School of Medicine at St. Mary’s, Immunology, London, U.K.). Clone 94.2 (YLCAWTGDQPQH) shared substantial homology with a T cell clone (YLCAWSGTSNQPQH) previously reported to the GenBank (GenBank GI no. 4038115; provided by B. J. Manfras, University Hospital Ulm, Ulm, Germany).
Discussion
We report in this work that T cells infiltrating the affected skin of patients with SSc contain substantial proportions of clonally expanded T cells. These findings demonstrate that only few β-chain TCR families were used by these infiltrating T cells, although it is likely that other less abundant β-chain TCR transcripts using additional Vβ segments may have not been detectable on agarose gel electrophoresis due to the low number of PCR cycles used in the nested PCR. Although a single TCR transcript that was clonally expanded in all patients was not identified, large proportions of β-chain TCR transcripts were identical in each one of the five patients. In particular, β-chain TCR transcripts exhibited a high degree of clonality ranging from 43 to 90%. The S70, S94, S169A, S169B, and S169C skin biopsies contained monoclonal expansions of T cells. Patients S162 and S168 showed oligoclonal expansions (more than one) of T cells. The fact that these clonal expansions were identified using two different amplification methods, the NPA-PCR (biopsy specimens S70, S94, S169A, and S169B) and the Vβ-specific PCR (biopsy specimens S162, S168, and S169C) enhances the validity of these findings. The oligoclonality of T cells that we identified in skin biopsies from patients with SSc is further supported by the detection in one patient (S169) of the same dominant T cell clone in three skin biopsies obtained at different time points and from three different skin regions (S169A, S169B, and S169C). These clonal expansions were identified in specimen S169C by Vβ-specific PCR and in specimens S169A and S169B by NPA-PCR. These findings indicate that T cells infiltrating the skin of patients with SSc have undergone in situ mono/oligoclonal expansion in response to an as yet unidentified Ag(s).
Restricted TCR CDR3 lengths were detected in CD8+ T cells from the bronchoalveolar lavage from patients with SSc, suggesting oligoclonal expansion of T cells (54). A restricted usage of TCR Vβ genes was also found in CD4−CD8− T cells from the peripheral blood of patients with SSc (55). In Tsk2 mice, an experimental model of SSc, it was also found that T cells infiltrating the affected skin of patients with SSc used restricted TCR Vβ gene segments (56). Apart from αβ TCR+ T cells, γδ TCR+ T cells were found to display evidence of activation and oligoclonal expansion. Increased proportion of circulating Vδ1 T cells expressed activation Ags (HLA-DR and CD49d) (57) and accumulated in the perivascular areas in affected SSc dermis (57). Furthermore, the TCR Vδ1 junctional region lengths were found to be skewed, which was suggestive of oligoclonal expansion of γδTCR+ T cells (58), and this observation was confirmed by sequence analysis (58). Comparison of the T cell populations present in the peripheral blood to those infiltrating skin lesions of patients with SSc is a different question from the one addressed in this work and will be the subject of another study. We do not know whether the α- and β-chain clonal expansions of T cells that we identified in skin biopsies from patients with SSc are restricted in the skin only, or whether they are also present in the peripheral blood of these patients. It is possible that the nature of the Ags that are driving these clonal expansions, i.e., whether they are alloantigen(s) or autoantigen(s), plays a role in determining their presence.
Previous studies have shown an association of particular HLA alleles with SSc (59, 60, 61, 62, 63, 64, 65, 66, 67), which supports the concept that an Ag-driven T cell response is important in the pathogenesis of SSc. Both an SSc-susceptible Caucasian population (65) and an SSc-susceptible Native American population (66) demonstrated a significant association of HLA-DQA1*0501 with the disease. However, analysis of five families with multicase SSc cases revealed that genes within the HLA system are required but are not sufficient to confer SSc (68). Lambert et al. (64) reported that persistent fetal microchimerism in T cells is associated with HLA-DQA1*0501 in a study with a limited number of patients (64). However, Artlett et al. (69) in a larger study found that the DQA1*0501 allele is not necessary for the establishment of microchimerism in the affected mother or fetus. The DQA1*0501 allele appears to be a risk factor for the development of microchimerism in idiopathic inflammatory myopathies (69).
Our results strongly suggest that T cells infiltrating the affected skin of patients with SSc have undergone proliferation and clonal expansion in response to a specific Ag(s). Although these Ag(s) are not known, several putative SSc Ags have emerged, including those described in the following sections.
Allogeneic cells and/or HLA of fetal or maternal origin
As previously discussed in the introduction of this paper, fetal cells have been identified in the peripheral blood and the skin of women with SSc who had been previously pregnant (14, 15, 16, 17, 18, 19, 20). Similarly, maternal cells have been identified in the peripheral blood or skin in women with SSc who have not been previously pregnant or in men with SSc (20, 21, 22, 23, 24). It has been proposed that SSc is caused, respectively, by a fetal antimaternal GVHD or by a maternal anti-offspring GVHD (18, 20). This is a novel and interesting concept and fits nicely with the clinical manifestations and serological profile of patients with GVHD (22). Fibrosis, microvascular fibrointimal proliferation, and anti-topoisomerase I autoantibodies have been described both in GVHD and in SSc (22). Activation of the fetal T cells present in the mother or of the maternal T cells present in the offspring, in response to an unknown antigenic stimulation, environmental factors such as viruses or chemicals, or because of breaking of the tolerance mechanisms that regulate the coexistence/cohabitation of fetal/maternal or maternal/offspring hemopoietic cells may be responding for the initiation of the disease. In this event, an allospecific T cell response, either of fetal T cells to maternal alloantigen(s) or of maternal T cells to offspring alloantigens, will be responsible for the initiation and likely the propagation of SSc.
Cytomegalovirus
CMV has been proposed as an agent that may participate in the pathogenesis of SSc in view of the increased levels of anti-CMV Abs in patients with SSc and the remarkable similarities between CMV vasculopathy and SSc vascular changes (70).
Retroviruses
Environmental factors, such as retroviruses (70, 71), may contribute to an increased frequency of cancer in first-degree relatives of patients with SSc (72).
DNA topoisomerase I and other autoantigens
Autoantibody responses to a number of Ags have been described in patients with SSc (73). Among autoantibodies frequently detected in patients with SSc are anti-topoisomerase I, anti-centromere, anti-fibrillin, and anti-RNA polymerases I, II, and III (73). Abs to a centromere kinesin-like protein are found in a limited form of SSc known as CREST and are associated with particular HLA-DQB1 alleles (74, 75). Anti-DNA topoisomerase I autoantibodies are perhaps the most prominent. It has long been known that T cells help B cells in Ab production and it has been confirmed that T cells are necessary for anti-topoisomerase I Ab production (13). T cells reactive to DNA topoisomerase I exhibit highly restricted TCR CDR3 (76). However, identical TCR CDR3 sequences were used by T cells reactive to DNA topoisomerase I from healthy donors (76). Therefore, the anti-topoisomerase I Ab response may be autoimmune, and may take place because of the extensive disregulation of the immune system because of the disease. The possibility that an autoantigen, which is the target of autoantibodies, also drives the clonal T cell expansion in SSc may be supported by the findings that the association of HLA is stronger with autoantibody profiles rather than the disease itself (59, 62, 76). Other putative autoantigens include fibrillarin, proteins of the centromere/kinetochore, and several proteins of the nucleolus (74, 75, 77). Autoantibodies to these Ags are found in small proportions of patients with SSc. Patients with particular HLA-DQ and -DR alleles were associated with higher-serum anti-topoisomerase Abs levels, and immunoreactivities to DNA topoisomerase I were associated with these alleles (63). Certain HLA-DRB1 or DQB1 alleles are associated with high-serum anti-centromere Ab (ACA) titers (78).
The finding of oligoclonal expansion of T cells in SSc reinforces the concept of T cell involvement in SSc and renders T cells a therapeutic target in SSc, a disease for which currently available therapies are largely ineffective. Depletion of CD4 T cells has already been performed in two patients with SSc and was associated with clinical improvement (79). Another approach might be the redirection of cytokines produced by activated T cells toward Th1 (80, 81). Anti-IL-4 Ab or soluble IL-4R alone or in combination with IFN-α or IL-12 are attractive candidate agents in this respect (81, 82, 83, 84); stimulatory molecules CD40 ligand and CD28/CTLA-4 are other possible targets.
In conclusion, we report the presence of strong mono/oligoclonal expansions of T cells infiltrating the affected skin of all five patients with SSc of recent onset examined in this study. Although it is clear that these clonal expansions must be Ag driven, the identity of the Ag(s) eliciting these responses is not known. It remains to be determined whether these Ag(s) are alloantigens or autoantigens. Identifications of these Ag(s) will significantly improve our understanding of the pathogenesis of the disease and will permit the development of new molecular and cellular approaches for the treatment of SSc.
Footnotes
-
↵1 This work was supported in part by National Institutes of Health Grants R01 AR48042 and T32 AI07101 (to C.D.P.), and Grant RO1 AM19616 (to S.A.J.).
-
↵2 L.I.S. and B.X. contributed equally to this manuscript.
-
↵3 Current address: Department of Medicine/Rheumatology, Thessaly University School of Medicine, Larisa, Greece.
-
↵4 Address correspondence and reprint requests to Dr. Chris D. Platsoucas, Department of Microbiology and Immunology, Temple University School of Medicine, 3400 North Broad Street, Philadelphia, PA 19140. E-mail address: cplatsoucas{at}vm.temple.edu
-
↵5 Abbreviations used in this paper: SSc, systemic sclerosis; NPA, nonpalindromic adaptor; GVHD, graft-vs-host disease; hpf, high-power field; ACA, anti-centromere Ab.
- Received March 21, 2001.
- Accepted January 25, 2002.
- Copyright © 2002 by The American Association of Immunologists
References
In this issue
