Abstract
Increased amounts of anti-neutrophil cytoplasm antibody (ANCA) directed against proteinase 3 (PR3) are a diagnostic and pathogenic hallmark of full-blown Wegener's granulomatosis (WG). Aggregates of B lymphocytes proximal to PR3+ cells as well as plasma cells have been described as substantial components of Wegener's granuloma and could participate in forming tertiary lymphoid structures, which might promote autoantibody formation. Our aim was a molecular analysis of single B cells in order to develop a methodological approach that allows examination of potential ANCA formation in the tissue. Single B cells from cryo-conserved endonasal biopsies of three WG patients were isolated, using laser-assisted microdissection. Subsequently, their immunoglobulin variable heavy (VH) and light (Vκ, Vλ) chain genes were analysed by single cell polymerase chain reaction and direct sequencing. Sixteen immunoglobulin VH-Vκ or VH-Vλ chain gene couples were characterized. Twelve of these immunoglobulin gene couples resembled memory B cells. Two offsprings of one B cell were detected, indicating clonal expansion. VH genes representing 39 single B cells of WG tissues displayed significantly more mutations when compared with VH genes from peripheral blood of a healthy donor. The findings confirm and extend our previous results, arguing for an initial selection and affinity maturation of B cells within Wegener's granuloma. Further, the methodology provides the initial basis for the recombinant generation of antibodies derived from tissue cells.
Keywords: B lymphocyte, immunoglobulins, laser microdissection, PCR, Wegener's granulomatosis
Introduction
Chronic inflammation of various tissues as well as uncontrolled autoantibody production are thought to be central pathogenetic mechanisms in autoimmune diseases. Animal models of autoimmune diabetes have underscored the importance of tertiary lymphoid structures as a promoter of autoreactive B cell selection as well as of target tissue inflammation[1]. Wegener's granulomatosis (WG) is characterized by such a chronic tissue inflammation as well as by systemic vasculitis, affecting predominantly small vessels [2,3]. Anti-neutrophil cytoplasm antibodies (ANCA) targeting one distinct autoantigen, ‘Wegener's autoantigen’ proteinase 3 (PR3), are highly specific for WG [4]. In vitro and in vivo studies support a pivotal role of ANCA in the induction of autoimmune vasculitis [5,6]. However, WG often begins as localized disease with granulomatous inflammation of the respiratory tract without circulating ANCA [7,8]. While the role of ANCA in the induction of vasculitis has been investigated extensively, much less is known about the autoantibody's origin and the process of B cell selection and maturation involved in its formation. We demonstrated that granulomas of the upper respiratory tract are infiltrated by clusters of B lymphocytes in vicinity to PR3+ cells, dendritic cells and plasma cells. Further, we found signs of potentially antigen (PR3)-driven selection within the immunoglobulin (Ig) heavy (VH) chain gene repertoire of seven different WG tissues [9–11]. Complete remission of refractory WG which has been achieved by targeting B lymphocytes with rituximab [12], including a disappearance of ANCA, could be taken as an argument for pathogenic relevance of B cells. None the less, it remains enigmatic if initial ANCA formation could take place in the granulomatous lesions. A study on myasthenia gravis demonstrated evidence for an ongoing antigen-driven B cell proliferation and selection in thymic germinal centres, providing a source for acetylcholine receptor-specific plasma cell precursors [13].
In order to obtain tools for investigating the structure–function relationship of B cell selection in Wegener's granuloma, single B cells were picked from assumed pathological structures within three endonasal biopsies of WG patients and examined for their antibody-encoding heavy and light (Vκ, Vλ) chain genes. Laser-assisted microdissection (LMD) is a newer method permitting the molecular characterization of even single cells from stained tissues [14]. Polymerase chain reaction (PCR) of single B cells [15] enables analysis of Ig heavy and light chain genes. Our data characterize Ig genes of single B cells from WG tissues as the necessary prerequisite to generate recombinant antibodies.
Materials and methods
Patients
After obtaining the patient's written consent according to the Declaration of Helsinki, endonasal biopsy specimens were taken from three WG patients. The study design has been approved by the local ethics committee (# 07-058). Patients’ characteristics at the time of biopsy are described in Table 1. Histopathological diagnosis was performed by K. H.-U. The patients fulfilled the classification criteria for WG [3,16].
Table 1.
Patient characteristics.
| Patient no. (age at biopsy, years) | WG manifestation | WG disease duration | Histological analysis | WG disease activity at biopsy (BVAS 1) | CRP (mg/dl) at biopsy | cANCA at biopsy | Therapy at biopsy |
|---|---|---|---|---|---|---|---|
| 1 (28) | Localized (E) | 3 months | Granuloma, vasculitis | Active (3) | Negative | Negative | None |
| 2 (68) | Localized (E) | 19 years | Granuloma, necrosis | Smouldering (1) | < 3 | Negative | Cotrimoxazole |
| 3 (49) | Generalized (E, L, Ey, C, B) | 1 year | Granuloma | Active (12) | 4,16 | 1:256 | Cyclo-phosphamide, steroids |
Wegener's granulomatosis (WG) manifestation is classified as follows: E, ear, nose and throat; L, lung; Ey, eye; C, central nervous system involvement; K, kidney; B, systemic inflammation. Granuloma includes epitheloid cells, multi-nucleated giant cells, lymphoid infiltrates [2]. Anti-neutrophil cytoplasm antibodies (cANCA) immunofluorescence results are given in titres. BVAS, Birmingham Vasculitis Activity Score; CRP, C-reactive protein.
Tissue B cell staining
Biopsy specimen of cases 1–3 were snap-frozen in liquid nitrogen immediately after extraction and stored at −80°C. Palm membrane-stringed slides (Carl Zeiss, Göttingen, Germany) were irradiated for 1 h by ultraviolet (UV) light. Five to 7 µm cryosections were cut and mounted onto these slides. A brief fixing in acetone was followed by a 15-min staining step with anti-CD20cy/horseradish peroxidase (Dako, Hamburg, Germany) and three Tris-buffered saline (TBS) (Tris/NaCl, pH 7·5) washes. Thereafter, the staining was developed for 5 min using 3-amino-9-ethyl-carbazole (Dako) followed by another TBS wash and haematoxylin counterstaining.
Laser-assisted microdissection
The Palm Microbeam system (Carl Zeiss) was employed following the manufacturer's instructions. Briefly, the cap of a laser pressure catapulting (LPC) microtube (Carl Zeiss) was coated with mineral oil (Sigma-Aldrich, Munich, Germany) and the microtube was placed into the clamp of the Axiovert 200 microscope. Using Palm Robo software (version 2·2) the marked cell was isolated automatically by cutting it out with a UV-A laser and catapulting it from the slide into the cap of the microtube. Afterwards, 20 µl of phosphate-buffered saline were added into the LPC microtube, followed by a short centrifugation and subsequent storage at −20°C.
Semi-nested single cell PCR
The semi-nested single cell PCR protocol for Ig VH and Vκ- or Vλ-chain genes consisted of two rounds of PCR, according to Kueppers and colleagues [15,17]. The DNA of the isolated cell was amplified in the first round with a mix of 29 primers representing Ig DNA sequences. PCR products from the first round were subjected consecutively to second rounds of PCR reactions employing individual VH-, Vκ- and Vλ-specific 5′ oligonucleotide primers and a mix of JH-, Jκ-, Jλ-specific 3′ oligonucleotide primers for the joining segments (Table 2). One of seven PCR reactions was run without DNA in the first and the second rounds as a negative control. All PCR control reactions without DNA remained negative. PCR was performed using the T3 Thermocycler (Biometra, Göttingen, Germany), according to the conditions described previously [15,17]. PCR products were analysed by gel electrophoresis on 1·5% agarose and ethidium bromide staining.
Table 2.
Oligonucleotide sequences.
| VH | |
| VH1 | 5′-CAG TCT GGG GCT GAG GTG AAG A-3′ |
| VH2 | 5′- GTC CTR CGC TGG TGA AAC CCA CAC A-3′ |
| VH3 | 5′-GGG GTC CCT GAG ACT CTC CTG TGC AG-3′ |
| VH4 | 5′-GAC CCT GTC CCT CAC CTG CRC TGT C-3′ |
| VH5 | 5′-AAA AAG CCC GGG GAG TCT CTG ARG A-3′ |
| VH6 | 5′-ACC TGT GCC ATC TCC GGG GAC AGT G-3′ |
| 1st round | |
| JH Intron 1 | 5′-GAC TCA CCT GAG GAG ACG GTG ACC-3′ |
| JH Intron 2 | 5′-TCT TAC CTG AGG AGA C GG TGA C CR T-3′ |
| 2nd round | |
| JH1 | 5′-GAC CAG GGT GCC CTG GCC CCA GTG C-3′ |
| JH2 | 5′-GAC CAG GGT GCC ACG GCC CCA GAG A-3′ |
| JH3 | 5′-CAT TGT CCC TTG GCC CCA GAC ATC A-3′ |
| JH4 | 5′-GAC CAC GGT TCC TTG GCC CCA GTA G-3′ |
| JH5 | 5′-GTG ACC AGG GTT CCT TGG CCC CAG G-3′ |
| JH6 | 5′-CGT GGT CCC TTG CCC CCA GAC GTC C- 3′ |
| Vκ | |
| Vκ1 | 5′-GAC ATC CRG WTG ACC CAG TCT CCW TC-3′ |
| Vκ2 | 5′-CAG WCT CCA CTC TCC CTG YCC GTC A-3′ |
| Vκ3 | 5′-TTG TGW TGA CRC AGT CTC CAG SCA CC-3′ |
| Vκ4 | 5′-AGA CTC CCT GGC TGT GTC TCT GGG C-3′ |
| Vκ5 | 5′-CAG TCT CCA GCA TTC ATG TCA GCG A-3′ |
| Vκ6 | 5′-TTT CAG TCT GTG ACT CCA AAG GAG AA-3′ |
| 1st round | |
| 3′Jκ1.2.4 | 5′-ACT CAG GTT TGA TYT CCA SCT TGG TCC-3′ |
| 3′Jκ3 | 5′-GTA CTT ACG TTT GAT ATC CAC TTT GGT CC-3′ |
| 3′Jκ5 | 5′-GCT TAC GTT TAA TCT CCA GTC GTG TCC-3′ |
| 2nd round | |
| 5′Jκ1.2. | 5′-TTG ATY TCC ASC TTG GTC CCY TGG C-3′ |
| 5′Jκ3 | 5′-TTG ATA TCC ACT TTG GTC CCA GGG C-3′ |
| 5′Jκ4 | 5′-TTG ATC TCC ACC TTG GTC CCT CCG C-3′ |
| 5′Jκ5 | 5′-TTA ATC TCC AGT CGT GTC CCT TGG C-3′ |
| Vλ | |
| Vλ1 | 5′-GGT CCT GGG CCC AGT CTG TG-3′ |
| Vλ2 | 5′-CAG TCT GCC CTG ACT CAG CCT-3′ |
| Vλ3a | 5′-CTC AGC CAC CCT CAG TGT CCG T-3′ |
| Vλ3b | 5′-CTC AGC CAC CCT CGG TGT CAG T-3′ |
| Vλ4 | 5′-TTT CTT CTG AGC TGA CTC AGG AC-3′ |
| Vλ6 | 5′-GAG TCT CCG GGG AAG ACG GTA-3′ |
| Vλ7 | 5′-GTG GTG ACT CAG GAG CCC TCA C-3′ |
| Vλ8 | 5′-ACT GTG GTG ACC CAG GAG CCA-3′ |
| Vλ9 | 5′-GCT GAC TCA GCC ACC TTC TGC A-3′ |
| 1st round | |
| 3′Jλ1 | 5′-GCC ACT TAC CTA GGA CGG TGA C-3′ |
| 3′Jλ2.3 | 5′-GAA GAG ACT CAC CTA GGA CGG TC-3′ |
| 3′Jλ6.7 | 5′-GGA GAC TYA CCG AGG ACG GTC-3′ |
| 2nd round | |
| 5′Jλ1 | 5′-GGA CGG TGA CCT TGG TCC CAG T-3′ |
| 5′Jλ2.3.7 | 5′-GAC GGT CAG CTT GGT SCC TCC-3′ |
| 5′Jλ6 | 5′-GACGGT CAC CTT GGT GCC ACT-3′ |
Sequencing and mutation analysis
The PCR products were purified and subjected to direct automatic sequencing (MWG-Biotech AG, Martinsried, Germany) using the same 5′ and 3′ primers as employed in the respective PCR. To assess if the mutational pattern reflects an antigen-driven process, a ratio of equal to or more than 2.9 regarding the ratio between amino acid replacement versus silent mutations (R : S ratio ≥ 2·9) within the complementarity determining region (CDR) was applied [18]. To characterize a B cell as memory type an R : S ratio ≤ 1·5 within the framework region (FR) of the Ig genes was employed [19].
Data from peripheral VH genes of a healthy volunteer
The sequences of 84 peripheral VH genes from a healthy volunteer [20] were extracted from the National Center for Biotechnology Information nucleotide database. Each gene sequence was analysed using the same approach as for the genes found in this study. These data, from a healthy donor's peripheral single B lymphocytes, were taken as control because of their comparable methodology.
Statistical analysis
Functional genes only, i.e. genes with a reading-frame that potentially encodes functional Ig chains, were analysed. Statistical analysis was performed as reported previously [9] using spss software (version 15.0; SPSS, Inc., Chicago, IL, USA).
Results
Gene characteristics of 16 single B cells bearing coding heavy and light chains
In the case of 16 laser-microdissected cells we detected B cell receptor genes encoding either a VH-Vκ or a VH-Vλ couple: seven cells displayed Vκ and nine cells exhibited Vλ chains. Five pairs were found in B cells from patient 1, eight pairs were isolated from B cells of patient 2 and three pairs were obtained from B cells of patient 3. Within the VH genes, the VH4 family was represented with seven mutated rearranged genes followed by VH1 with six and VH3 with three. From seven κ-light chains two each were Vκ1, Vκ2 and Vκ3 respectively, and one was Vκ5. From nine λ-light chains four were Vλ1, three were Vλ3, one each was Vλ2 and Vλ6 respectively (Table 3).
Table 3.
Sixteen VH/Vκ and VH/Vλ gene couples derived from single B cells of three Wegener's granulomatosis (WG) tissue samples.
| Patient no. | Sequence | Germline gene | Homology (% to germline) | CDR | FR | CDR 3 length | ||
|---|---|---|---|---|---|---|---|---|
| R | S | R | S | |||||
| 1 | Hm0507-10-VH4 | 4-39 | 89 | 6 | 2 | 7 | 5 | 16 |
| Hm0507-10-VK2 | 2-28 | 99 | 0 | 0 | 1 | 0 | 7 | |
| 1 | Hm0207-17-VH1 | 1-69 | 99 | 0 | 0 | 1 | 1 | 12 |
| Hm0207-17-VL1 | 1-74 | 99 | 3 | 0 | 6 | 2 | 13 | |
| 1 | Hm0407-08-VH4 | 4-39 | 90 | 5 | 1 | 9 | 6 | 18 |
| Hm0407-08-VL1 | 1-44 | 94 | 3 | 2 | 8 | 4 | 11 | |
| 1 | Hm0507-12-VH1 | 1-46 | 93 | 4 | 2 | 4 | 3 | 17 |
| Hm0507-12-VL1 | 1-40 | 98 | 2 | 0 | 2 | 0 | 11 | |
| 1 | Hm1207-35-VH3 | 3-21 | 92 | 3 | 0 | 6 | 2 | 20 |
| Hm1207-35-Vκ3 | 3-11 | 98 | 0 | 0 | 2 | 0 | 9 | |
| 2 | Scu3-VH4 | 4-4 | 90 | 6 | 0 | 5 | 9 | 11 |
| Scu3-VL6 | 6-57 | 99 | 1 | 0 | 1 | 0 | 9 | |
| 2 | Scu68-VH4 | 4-61 | 85 | 8 | 4 | 12 | 7 | 17 |
| Scu68-VL3 | 3-19 | 85 | 0 | 0 | 4 | 1 | 12 | |
| 2 | Scu0107-36-VH4 | 4-31 | 96 | 1 | 1 | 7 | 0 | 13 |
| Scu0107-36-VK1 | 1-16 | 97 | 1 | 1 | 2 | 1 | 9 | |
| 2 | Scu0107-58-VH3 | 3-30 | 88 | 5 | 3 | 9 | 3 | 21 |
| Scu0107-58-VK3 | 3-15 | 91 | 3 | 1 | 11 | 4 | 10 | |
| 2 | Scu0107-42-VH1 | 1-18 | 98 | 2 | 0 | 1 | 2 | 17 |
| Scu0107-42-VL1 | 1-36 | 98 | 0 | 0 | 2 | 2 | 11 | |
| 2 | Scu0107-45-VH1 | 1-18 | 89 | 6 | 1 | 12 | 5 | 15 |
| Scu0107-45-VL2 | 2-14 | 98 | 3 | 0 | 0 | 0 | 10 | |
| 2 | Scu0107-53-VH1 | 1-3 | 99 | 1 | 0 | 0 | 0 | 17 |
| Scu0107-53-VL3b | 3-1 | 96 | 3 | 1 | 3 | 2 | 9 | |
| 2 | Scu0107-59-VH1 | 1-18 | 92 | 2 | 1 | 10 | 2 | 19 |
| Scu0107-59-VL3b | 3-1 | 96 | 4 | 0 | 1 | 3 | 9 | |
| 3 | ST140-VH4 | 4-59 | 89 | 3 | 0 | 10 | 6 | 9 |
| ST140-VK1 | 1-12 | 100 | 0 | 0 | 0 | 0 | 10 | |
| 3 | ST146-VH4 | 4-34 | 92 | 2 | 2 | 9 | 3 | 14 |
| ST146-VK2 | 2D-40 | 99 | 0 | 0 | 1 | 0 | 10 | |
| 3 | ST177-VH3 | 3-48 | 86 | 0 | 0 | 7 | 4 | 17 |
| ST177-VK5 | 5-2 | 99 | 0 | 0 | 1 | 0 | 11 | |
Germline gene and homology indicate the percentage of identical nucleic acids compared with the respective germline gene. The total number of replacement (R) and silent (S) mutations within the complementarity determining region (CDR) and the framework region (FR) are listed. The CDR3 length is given in amino acid numbers.
Twelve of the 16 Ig gene couples revealed mutational patterns reflecting an antigen-driven process as well as characteristics of memory B cells [18,19,21]. Five of the Ig gene couples exhibited R : S ratios indicative of selection within both regions, CDR and FR. One of the Ig gene couples displayed an R : S ratio ≥ 2·9 within the CDR for both segments, VH and Vλ. Altogether, R : S ratios ≥ 2·9 were found for six heavy and five light chains of Ig gene couples. R : S ratios = 1·5 within the FR were observed for six heavy and three light chain genes. Of note, the light chain genes seem to be less mutated than the VH genes; nine light chain genes showed ≤ 1 replacing mutation within the CDR (Table 3).
Except for one couple, the CDR3 of all other heavy chain segments contained aspartic acid at position 116, which might be a hint towards selection against a positively charged molecule such as PR3.
Analysis of 39 VH genes derived from single B cells of three endonasal tissues
In addition to the above-described 16 cells, we detected functionally rearranged VH genes of another 23 single B cells isolated from endonasal biopsies of three WG patients (Table 4). We compared the mutational pattern of VH genes from single B cells of each WG patient to 71 VH genes from peripheral blood B lymphocytes of a healthy volunteer [20]. The mean mutation frequency was 6·3% for patient 1, 7·6% for patient 2 and 7·8% for patient 3, compared with 2·6% for the healthy donor. These differences were statistically significant. In terms of family distribution we observed that 16 genes belonged to the VH4, 14 genes to the VH1 and nine genes to the VH3 family. The VH4-34 segment was found six times and was the most-represented VH gene within the 39 WG-derived genes. Further, we calculated mean R : S ratios of the single cell VH genes for each WG patient. The R : S values for patient 1 (10 VH genes) were 3·9 for the CDR and 1·75 for the FR respectively. Such a distribution is typical for selection by antigen [18,19,21], especially in patient 1 with active endonasal disease. For patient 2 (11 VH genes) the mean R : S value was 2·7 for the CDR and and 1·8 for the FR, suggesting random mutation. For patient 3 (18 VH genes) the mean R : S ratio was 3·6 for the CDR and 2·1 for the FR, which is in line with an antigen-driven mutation. This patient with aggressive systemic disease displayed endonasal activity at the time of biopsy despite ongoing cyclophosphamide treatment.
Table 4.
Thirty-nine variable heavy (VH) sequences grouped according to the patient number.
| Patient no. | Sequence | VH germline gene | Homology (% to germline) | CDR | FR | CDR 3 length | ||
|---|---|---|---|---|---|---|---|---|
| R | S | R | S | |||||
| 1 | Hm0207-17-VH1 | 1-69 | 99 | 0 | 0 | 1 | 1 | 12 |
| 1 | Hm0207-25-VH3 | 3-33 | 95 | 4 | 1 | 3 | 2 | 12 |
| 1 | Hm0407-08-VH4 | 4-39 | 90 | 5 | 1 | 9 | 6 | 18 |
| 1 | Hm0407-11-VH4 | 4-34 | 100 | 0 | 0 | 0 | 0 | 19 |
| 1 | Hm0407-18-VH4 | 4-31 | 87 | 2 | 0 | 9 | 1 | 15 |
| 1 | Hm0507-10-VH4 | 4-39 | 89 | 6 | 2 | 7 | 5 | 16 |
| 1 | Hm0507-12-VH1 | 1-46 | 93 | 4 | 2 | 4 | 3 | 17 |
| 1 | Hm0707-05-VH3 | 3-30 | 93 | 3 | 1 | 2 | 4 | 18 |
| 1 | Hm1107-48-VH4 | 4-59 | 99 | 0 | 0 | 1 | 0 | 13 |
| 1 | Hm1207-35-VH3 | 3-21 | 92 | 3 | 0 | 6 | 2 | 20 |
| 2 | Scu3-VH4 | 4-4 | 90 | 6 | 0 | 5 | 9 | 11 |
| 2 | Scu68-VH4 | 4-61 | 85 | 8 | 4 | 12 | 7 | 17 |
| 2 | Scu0107-36-VH4 | 4-31 | 96 | 1 | 1 | 7 | 0 | 13 |
| 2 | Scu0107-42-VH1 | 1-18 | 98 | 2 | 0 | 1 | 2 | 17 |
| 2 | Scu0107-45-VH1 | 1-18 | 89 | 6 | 1 | 12 | 5 | 15 |
| 2 | Scu0107-50-VH1 | 1-2 | 94 | 2 | 3 | 3 | 3 | 14 |
| 2 | Scu0107-53-VH1 | 1-3 | 99 | 1 | 0 | 0 | 0 | 17 |
| 2 | Scu0107-58-VH3 | 3-30 | 88 | 5 | 3 | 9 | 3 | 21 |
| 2 | Scu0107-59-VH1 | 1-18 | 92 | 2 | 1 | 10 | 2 | 19 |
| 2 | Scu0507-07-VH1 | 7-4 | 90 | 1 | 0 | 8 | 7 | 19 |
| 2 | Scu0707-09-VH1 | 1-2 | 95 | 1 | 0 | 5 | 2 | 18 |
| 3 | ST140-VH4 | 4-59 | 89 | 3 | 0 | 10 | 6 | 9 |
| 3 | ST146-VH4 | 4-34 | 92 | 2 | 2 | 9 | 3 | 14 |
| 3 | ST155-VH1 | 1-3 | 90 | 2 | 1 | 7 | 5 | 18 |
| 3 | ST159-VH4 | 4-34 | 89 | 4 | 0 | 9 | 2 | 11 |
| 3 | ST177-VH3 | 3-48 | 86 | 0 | 0 | 7 | 4 | 17 |
| 3 | ST188-VH4 | 4-34 | 92 | 2 | 1 | 9 | 2 | 14 |
| 3 | ST210-VH1 | 1-46 | 92 | 0 | 0 | 4 | 4 | 11 |
| 3 | ST228-VH4 | 4-34 | 82 | 1 | 0 | 7 | 0 | 15 |
| 3 | ST231-VH1 | 1-3 | 90 | 0 | 0 | 5 | 5 | 18 |
| 3 | ST231-VH2 | 1-3 | 90 | 2 | 1 | 8 | 5 | 18 |
| 3 | ST245-VH3 | 3-21* | 95 | 4 | 1 | 2 | 1 | 17 |
| 3 | ST248-VH3 | 3-48* | 98 | 0 | 0 | 2 | 1 | 17 |
| 3 | ST248-VH4 | 4-34 | 92 | 2 | 1 | 8 | 3 | 14 |
| 3 | ST253-VH3 | 3-30 | 98 | 1 | 1 | 1 | 0 | 17 |
| 3 | ST262-VH1 | 1-69 | 93 | 2 | 0 | 4 | 4 | 10 |
| 3 | ST273-VH2 | 4-39 | 96 | 2 | 0 | 3 | 1 | 16 |
| 3 | ST273-VH3 | 3-48 | 99 | – | – | 1 | 0 | 17 |
| 3 | ST273-VH4 | 4-39 | 96 | 2 | 0 | 3 | 1 | 16 |
St 245 and 248 represent differently mutated offsprings of one identically rearranged VH gene. The different reference germline genes can be explained through the mutations within CDR/FR 1 and 2. CDR, complementarity determining region; FR, framework region; R, replacement; S, silent mutations.
Clonal expansion
The biopsy section of patient 3 contained two separately analysed single B cells (St245 and St248) which carried the same rearranged VH3-21 gene. This is an example of clonal expansion. Interestingly, the CDR3 (joining segment) featured six negatively charged amino acid residues (E, D) that, potentially, favour affinity to the positively charged PR3.
Discussion
In order to examine individual B cell receptors from WG, our study focused on single B cells derived from endonasal granulomas. Three distinct WG manifestations, one early-onset active localized WG, one relapsed localized WG and one full-blown systemic WG, were compared. For the first time we describe light and heavy chain Ig gene couples of 16 single B lymphocytes isolated from WG tissues. Seventy-five per cent (i.e. 12 of 16) of the Ig gene couples displayed signs of B cell selection and maturation characterized by R : S ratios = 2·9 within the CDR and = 1·5 within the FR [18,19]. Previously, we demonstrated B lymphocyte maturation in Wegener's granuloma, based upon a comparison of VH genes from six endonasal and one lung biopsies to healthy controls and concluded a local antigen-driven process in active WG [9,10]. However, these results as well as a large VH gene analysis from thymic germinal centres in myasthenia gravis [13] were not based on single cells and thus did not allow further characterization of individual B cell receptors.
The sequences of single B cell-derived functional VH genes of each patient displayed a significantly higher mutation frequency compared with healthy control [20]. The different mutational patterns correlated with disease activity at the time of biopsy. Further, in accordance with previous results [9,10], there seems to be an association between active inflammation and selective pressure. The detection of distinct offsprings of one B lymphocyte clone from patient 3 indicated clonal expansion. Two separate CD20+ cells were isolated within an area of stained B cells in the tissue section and PCR products stemming from two separate reactions were obtained consecutively. Together with the differently mutated FR1 and CDR1 of the two clonally related cells, these findings argue against a contamination artefact. Interestingly, we observed clonal expansion in lung tissue of the same previous patient [10]. The healthy Ig heavy chain gene repertoire is dominated by VH3, followed by VH4 and VH1 genes [20]. The 39 VH genes represented these three families, but the order was different. Similar to our analysis of 184 VH genes from six endonasal lesions of WG [9], VH4-34 was the most frequently used gene segment (15%) from single B cells of WG tissues. B cells expressing this gene have been designated as inherently autoreactive [22]. These findings might be supportive of an autoimmune response, as the VH gene data from a healthy repertoire suggested a negative selection against autoantibody-associated VH4 genes [20]. A critical point of our study that we used control data derived from peripheral blood B cells instead of B cells from tissue. However, the latter are difficult to obtain for several reasons. A molecular analysis of nasal control tissue from a patient with conchal hyperplasia yielded no Ig genes [9], indicating that there were no B cells. When staining for CD20+ cells in nasal tissue of sinusitis patients, B cell aggregates were observed (unpublished data). However, sinusitis does not represent healthy/non-inflamed tissue but an inflammatory response, perhaps even similar to WG. Therefore, in our opinion the use of peripheral blood B cell data from a healthy individual is probably not the most accurate, but until the present is still the most feasible control. Altogether, the single B cell data emphasize and extend our previous data. None the less, we are well aware that our results are based upon a comparatively low number of B lymphocytes and patients. However, despite the use of contemporary techniques such as LMD and single cell PCR it still remains a challenge to obtain enough and clean material from single tissue lymphocytes [15,23]. Moreover, our initial approach aimed at laying a methodological foundation for the generation of recombinant Ig from tissue and was not directed primarily at analysing high numbers of B cells as previously [9]. In conclusion, our observations indicate selection and maturation of single, potentially PR3-reactive B cells, supporting the idea that an autoimmune response directed against PR3 may take place in Wegener's granuloma. To elucidate further the structure–function relationships of the mutated genes and its resulting Ig, we attempt to generate recombinant antibodies on the basis of the 16 Ig gene couples.
Acknowledgments
We thank Professor Dr Peter Lamprecht for critical reading of the manuscript, Dr Katrin Kallies, Institute of Anatomy, University of Luebeck, Germany for introduction in LMD techniques and Petra Zander for technical assistance. Funding was provided by the German Research Society (Clinical Research Unit 170TP4 to K. H.-U., W. L. G., A. M.) the Association for the advancement in research of rheumatic diseseases Bad Bramstedt e.V. (0·2 to A. M.) and the Faculty of Medicine of the University of Saarland (HOMFOR to J. V.).
References
- 1.Kendall PL, Yu G, Woodward EJ, Thomas JW. Tertiary lymphoid structures in the pancreas promote selection of B lymphocytes in autoimmune diabetes. J Immunol. 2007;178:5643–51. doi: 10.4049/jimmunol.178.9.5643. [DOI] [PubMed] [Google Scholar]
- 2.Müller A, Holl-Ulrich K, Lamprecht P, Gross WL. Germinal centre-like structures in Wegener's granuloma: the morphological basis for autoimmunity? Rheumatology. 2008;47:1111–3. doi: 10.1093/rheumatology/ken202. [DOI] [PubMed] [Google Scholar]
- 3.Jennette JC, Falk RJ, Andrassy K, et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum. 1994;37:187–92. doi: 10.1002/art.1780370206. [DOI] [PubMed] [Google Scholar]
- 4.Schönermarck U, Lamprecht P, Csernok E, Gross WL. Prevalence and spectrum of rheumatic diseases associated with proteinase 3-antineutrophil cytoplasmic antibodies (ANCA) and myeloperoxidase-ANCA. Rheumatology. 2001;40:178–84. doi: 10.1093/rheumatology/40.2.178. [DOI] [PubMed] [Google Scholar]
- 5.Falk RJ, Terrell RS, Charles LA, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Natl Acad Sci. 1990;87:4115–19. doi: 10.1073/pnas.87.11.4115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pfister H, Ollert M, Froehlich LF, et al. Antineutrophil cytoplasmic autoantibodies against the murine homolog of proteinase 3 (Wegener autoantigen) are pathogenic in vivo. Blood. 2004;104:1411–18. doi: 10.1182/blood-2004-01-0267. [DOI] [PubMed] [Google Scholar]
- 7.Reinhold-Keller E, Beuge N, Latza U, et al. An interdisciplinary approach to the care of patients with Wegener's granulomatosis. Arthritis Rheum. 2000;43:1021–32. doi: 10.1002/1529-0131(200005)43:5<1021::AID-ANR10>3.0.CO;2-J. [DOI] [PubMed] [Google Scholar]
- 8.Jayne D. Update on the European Vasculitis Study Group trials. Curr Opin Rheumatol. 2001;13:48–55. doi: 10.1097/00002281-200101000-00008. [DOI] [PubMed] [Google Scholar]
- 9.Voswinkel J, Müller A, Krämer JA, et al. B lymphocyte maturation in Wegener's granulomatosis: a comparative analysis of VH genes from endonasal lesions. Ann Rheum Dis. 2006;65:859–64. doi: 10.1136/ard.2005.044909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Krämer JA, Müller A, Herlyn K, et al. B lymphocyte differentiation in granulomatous tissues of the lung and the nasal mucosa in Wegener's granulomatosis: origin of anti-neutrophil cytoplasmic antibody formation? Z Rheumatol. 2007;66:421–9. doi: 10.1007/s00393-007-0170-8. [DOI] [PubMed] [Google Scholar]
- 11.Capraru A, Müller A, Csernok E, et al. Expansion of circulating NKG2D+ effector memory T-cells and expression of NKG2D-ligand MIC in granulomatous lesions in Wegener's granulomatosis. Clin Immunol. 2008;127:144–50. doi: 10.1016/j.clim.2007.12.004. [DOI] [PubMed] [Google Scholar]
- 12.Golbin JM, Specks U. Targeting B lymphocytes as therapy for ANCA-associated vasculitis. Rheum Dis Clin North Am. 2007;33:741–54. doi: 10.1016/j.rdc.2007.09.001. [DOI] [PubMed] [Google Scholar]
- 13.Sims GP, Shiono H, Willcox N, Stott DI. Somatic hypermutation and selection of B cells in thymic germinal centers responding to acetylcholine receptor in myasthenia gravis. J Immunol. 2001;167:1935–44. doi: 10.4049/jimmunol.167.4.1935. [DOI] [PubMed] [Google Scholar]
- 14.Fink L, Kwapiszewska G, Wilhelm J, Bohle RM. Laser-microdissection for cell type- and compartment-specific analyses on genomic and proteomic level. Exp Toxicol Pathol. 2006;57:25–9. doi: 10.1016/j.etp.2006.02.010. [DOI] [PubMed] [Google Scholar]
- 15.Küppers R. Molecular single cell PCR analysis of rearranged immunoglobulin genes as a tool to determine the clonal composition of normal and malignant human B cells. Meth. Mol. Biol. 2004;271:225–38. doi: 10.1385/1-59259-796-3:225. [DOI] [PubMed] [Google Scholar]
- 16.Leavitt RY, Fauci AS, Bloch DA, et al. The American College of Rheumatology 1990 criteria for the classification of Wegener's granulomatosis. Arthritis Rheum. 1190;33:1101–7. doi: 10.1002/art.1780330807. [DOI] [PubMed] [Google Scholar]
- 17.Bräuninger A, Küppers R, Spieker T, et al. Molecular analysis of single B cells from T-cell-rich B-cell lymphoma shows the derivation of the tumor cells from mutating germinal center B cells and exemplifies means by which immunoglobulin genes are modified in germinal center B cells. Blood. 1999;8:2679–87. [PubMed] [Google Scholar]
- 18.Chang B, Casali P. The CDR1 sequences of a major proportion of human germline Ig VH genes are inherently susceptible to amino acid replacement. Immunol Today. 1994;15:367–73. doi: 10.1016/0167-5699(94)90175-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Klein U, Goossens T, Fischer M, et al. Somatic hypermutation in normal and transformed human B cells. Immunol Rev. 1998;162:261–80. doi: 10.1111/j.1600-065x.1998.tb01447.x. [DOI] [PubMed] [Google Scholar]
- 20.Brezinschek HP, Brezinschek RI, Lipsky PE. Analysis of the heavy chain repertoire of human peripheral B cells using single-cell polymerase chain reaction. J Immunol. 1995;155:190–202. [PubMed] [Google Scholar]
- 21.Lossos IS, Tibshirani R, Narasimhan B, Levy R. The inference of antigen selection on Ig genes. J Immunol. 2000;165:5122–6. doi: 10.4049/jimmunol.165.9.5122. [DOI] [PubMed] [Google Scholar]
- 22.Pugh-Bernard AE, Silverman GJ, Cappione AJ, et al. Regulation of inherently autoreactive VH4-34 B cells in the maintenance of human B cell tolerance. J Clin Invest. 2001;108:1061–70. doi: 10.1172/JCI12462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Seitz S, Schneider CK, Malotka J, et al. Reconstitution of paired T cell receptor α- and β-chains from microdissected single cells of human inflammatory tissues. Proc Natl Acad Sci USA. 2006;103:12057–62. doi: 10.1073/pnas.0604247103. [DOI] [PMC free article] [PubMed] [Google Scholar]