European Journal of Immunology · 2014 · 28 citations · 3 references
Life SciencesCell BiologyImmunologyAutoimmunityCellular ImmunologyImmunotechnologyAntibody CharacterizationAntibody FunctionAntibody PurificationAntibody-antigen RecognitionComplement SystemHybridoma TechnologyAntibodiesAntibody Binding SpecificityAntigen Specificity DeterminationAntigen-specific Cell EnrichmentHybridoma Cell Line DevelopmentHybridoma Cell LinesSecreted AntigensSingle-cell Antibody Isolation
Plasma cells (PCs) are antibody-secreting cells (ASCs) that play an important role in immunity and autoimmunity. There is an urgent need to develop efficient and simple methods for analyzing the biology and function of PCs at the single-cell level. Here, we describe a method for the selective labeling of viable cells secreting antibodies of a given specificity. We demonstrate that these labeled PCs can be isolated alive or induced to commit suicide in the presence of complement. The method is based on affinity matrix (AM) technology that we developed previously 1, 2 and on a method to enrich mutants of a hybridoma cell line 3. Antigen specificity is provided by the conjugation of an antibody or a F(ab) fragment recognizing the surface markers CD138 or CD44 on PCs with the antigen of interest. The matrix is used to coat the surface of PCs with the antigen of interest — in this case, ovalbumin (OVA) or acetylcholine receptor (AChR). Secreted antibodies of the respective specificity bind to the antigen attached via the cell-surface AM of the secreting cells (Fig. 1A). The secreted antibody bound to the ASCs can then be identified and stained by immunofluorescence, using an anti-Ig isotype antibody. Alternatively, in the presence of complement, they can be induced to commit suicide, lysed by their own antibodies. The anti-CD138-OVA or anti-CD44-OVA AM strongly labels the surface of hybridoma cells, inducing an approximately 104 shift in mean fluorescence intensity when stained with antigen-specific antibody (Supporting Information Fig. 1). When the anti-OVA hybridoma cells are coated with the anti-CD138-OVA AM and incubated at 37°C for 10 min, the antigen in the matrix is efficiently labeled with secreted antibodies with a 58-fold increase in mean fluorescence intensity when the cells are stained extracellularly with fluorescent isotype-specific antibody (Supporting Information Fig. 2). The sensitivity of the method was analyzed by spiking populations of CFSE-labeled 4C9 hybridoma cells secreting OVA-specific antibodies into a population of X63-Ag6.653 myeloma cells at ratios of 1:1 to 1:10 000. The OVA-specific hybridoma cells were selectively labeled by secreted anti-OVA-IgG at all ratios analyzed (Fig. 1B). Decreasing the permeability of the incubation medium (see Supporting Information for details) limited the diffusion of the antibody of interest (cross-feeding); this was used to avoid unspecific labeling in culture, with a specific:unspecific cell ratio of 1:1 (Fig. 1C) and 2:1 (not shown). The AM also allowed the distinction of specific PCs with different IgG subclasses when hybridoma cells producing IgG1 or IgG2b anti-OVA antibodies were mixed (Supporting Information Fig. 3). Moreover, the AM can be used to rapidly select and sort viable anti-OVA-producing hybridoma cells in a cell line with partially lost OVA antibody production due to long-term culture, obviating the need for labor- and time-intensive subcloning and screening for reestablishment of the monoclonal culture (Supporting Information Fig. 4). This AM technology was also used ex vivo for the identification and sorting of viable antigen-specific PCs from the lymphoid organs of BALB/c mice immunized with OVA. Spleen and bone marrow PCs marked with the matrix (anti-CD138-OVA) were incubated for 30 min at 37°C to allow for antibody secretion. The secreted anti-OVA antibody captured by the matrix was labeled using an anti-mouse IgG1 antibody. As shown in Figure 1D, only OVA-specific PCs were labeled by the secreted antibody, as determined by the acquisition of IgG1 expression on their surfaces and by their intracellular OVA binding. Sorting the PCs based on the secreted IgG1 yielded an OVA-specific PC population with a purity >75% (Supporting Information Fig. 5). The AM technology was also used to selectively deplete PCs secreting antibodies of a given specificity with complement present in the culture medium (Fig. 2). As proof of concept, OVA-secreting hybridoma cells were incubated with the AM and cultured for 30 min at 37°C in medium supplemented with 10% guinea pig complement. After exclusion of dead cells by DAPI staining, viable cells were counted by flow cytometry. Anti-OVA-specific hybridoma cells were specifically lysed when the cells were coated with the matrix, while the CD138-positive X63-Ag6.653 myeloma cells were not significantly depleted (Fig. 2A) unless anti-OVA antibodies were added to the culture (Supporting Information Fig. 6). The specificity of the AM technology was confirmed by using a mixture of specific anti-OVA-secreting hybridoma cells and ASCs of unrelated specificity. As shown in Figure 2A, after incubation with the matrix OVA-specific ASCs were lysed to a greater extent than ASCs of unrelated specificity. Similar results were obtained by using a F(ab)-anti-CD44-OVA AM to deplete OVA-specific hybridoma cells (Supporting Information Fig. 7). Consequently, this AM technology is suited to deplete PCs ex vivo based on their antigen specificity. To demonstrate this, spleen and bone marrow cells from BALB/c mice subjected to secondary immunization with OVA were labeled with the AM and incubated in the presence of complement. After 30 min, OVA-specific PCs in the culture were enumerated by flow cytometry. As shown in Figure 2C, this significantly reduced the frequencies of viable anti-OVA-secreting PCs from the spleen. However, the numbers of PCs secreting antibodies against other antigens were not affected, confirming the specificity of the method for the antigen used in the matrix. Similar results were obtained using bone marrow PCs (Supporting Information Fig. 8). Next, we tested whether PCs secreting pathogenic autoantibodies can be lysed ex vivo using this AM technology. We established a murine model of experimental autoimmune myasthenia gravis 4 and constructed an AM conjugating the anti-CD138-antibody to the Torpedo-AChR, the autoantigen used to induce the experimental autoimmune myasthenia gravis 5. This matrix significantly reduced the frequency of splenic PCs secreting anti-AChR autoantibodies ex vivo (enumerated by their intracellular binding of Torpedo-AChR; Fig. 2D). The frequency of PCs secreting antibodies of other specificities did not change significantly, suggesting that the AM technology is a valid method for autoantigen-specific depletion of PCs. In conclusion, the AM technology described here is an innovative, inexpensive, and simple tool (i) for sorting and analyzing large numbers of (auto)antigen-specific-PCs, (ii) for reducing the time and manipulations necessary for hybridoma selection and cloning, and (iii) for the (auto)antigen-specific depletion of pathogenic memory PCs while leaving the protective PC memory intact. Its therapeutic efficacy in murine models of autoimmune diseases is currently under investigation. This work was supported by grants from the Deutsche Forschungsgemeinschaft (DFG) SFB650 TP17 and Transregio SFB TRR 130 TP15. We thank Robert Biesen, Department of Rheumatology and Clinical Immunology, Charité – Universitätsmedizin for assistance in data analyses and the DRFZ Flow Cytometry and Cell Sorting Facility for cell sorting. A patent application has been filed by the Deutsches Rheuma-Forschungszentrum (DRFZ), application number WO2014/037519 A2. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Supplementary Figure 1: The affinity matrix can bind the surface of ASC e myeloma cells. A) Hybridoma cells were incubated with the anti-CD138-OVA or Fab-anti-CD44-OVA affinity matrix (AM). An anti-OVA IgG-Cy5 antibody was used to detect the AM bound on the surface of the cells. Black line: Hybridoma cells labeled with the affinity matrix; grey line: hybridoma cells without AM. B) X63-Ag8.653 mouse Myeloma cells were incubated with the CD138-OVA and treated as described above. Shown are representative data from one of at least three independent experiments Supplementary Figure 2: The specific secreted antibody may bind to the affinity matrix (AM). Anti-OVA hybridoma cells were incubated (black line) or not incubated (grey line) with the AM. The cells were allowed to secrete for 10 min and the secreted antibody relocated to the cell surface was detected by an anti-mouse-IgG2b antibody. Shown are representative data from one of at least three independent experiments. Supplementary Figure 3:The affinity matrix (AM) technology allows the identification of subclassspecific antibody-secreting cells. A) Monoclonal cultures (first row) or mixed-populations (at ratios of 100:1 or 1000:1, second row) of anti-OVA-IgG2b hybridoma cells (clone SK19-2E8) and/or CFSEstained anti-OVA-IgG1 hybridoma cells (clone SK19-4C9) were incubated with the AM and cultured for 10 minutes of antibody secretion. The cells were stained with an anti-IgG1 antibody in order to identify the cells secreting IgG1-anti-OVA-antibodies in the mixed culture, and CFSE-staining was used to confirm that most of the CFSE+-IgG1 cells could be identified using the AM. Supplementary Figure 4 :The affinity matrix (AM) technology may be used to enrich OVA-specific cells in a line that has lost its OVA specificity due to long-term culture. Anti-OVA hybridoma cells with a partial loss of immunoglobulin production due to long-term culture were incubated with the AM, cultured for 10 minutes to allow for antibody secretion, stained with a biotinylated anti-IgG antibody, and sorted using anti-biotin microbeads and a MACS Separator. Unsorted cells and surface-IgGsorted cells were stained intracellularly with OVA-Cy5 to enumerate the percentage of OVA-specific cells. Shown are representative data from one of two independent experiments. Supplementary Figure 5: The affinity matrix (AM) technology can be used to sort OVA-specific cells from the spleen of immunized mice. A) Representative gating strategy for the sorting of cells secreting anti-OVA antibodies in the spleen of immunized mice. Spleen cells were labeled with the AM, allowed to secrete (see Methods for details), and stained extracellularly with anti-mouse-IgG1-PE-Cy7, anti-CD138-PE. A dump channel containing DAPI, CD3, CD11c, Gr-1, IgM, IgD was also included. Plasma cells were identified as DUMP-CD138+ cells. Among plasma cells, IgG1-positive (containing OVA-Agspecific cells) and -negative cells (PCs of other specificities) were sorted using a BD FACSAria cell sorter. B) To verify their antigen specificity, the sorted cells were intracellularly stained for kappa light chain-PO and OVA-FITC, and then analyzed by FACS. Supplementary Figure 6: Anti-OVA antibodies may induce the AM-mediated lysis of myeloma cells. X63-Ag8.653 myeloma cells were incubated or not incubated (controls) with the AM and cultured for 30-45 min (37°C, 5% CO2) in the presence of 10% guinea pig complement. An anti-OVA-IgG2b antibody was added to the culture where indicated. After exclusion of dead cells by DAPI staining, live cells were quantified by FACS. The data are mean ± SE of the mean of three replicates of a single experiment, and are representative of at least three independent experiments. *p ≤ 0.05 by Mann-Whitney U test. Supplementary Figure 7: Affinity matrix-mediated cells lysis is independent from the antibody used to target the ASC. Anti-OVA-IgG2b hybridoma cells (clone SK19-1E10) were incubated with or without (controls) with the F(ab)-fragment-anti-CD44-OVA affinity matrix, and then plated and cultured for 30 min in the presence of 10% guinea pig complement. An anti-OVA-IgG2b antibody was added to the culture where indicated. After exclusion of dead cell by DAPI staining, live cells were quantified by FACS. The data are mean ± SE of three replicates of a single experiment, and are representative of two independent experiments. *p ≤ 0.05 by Mann-Whitney U test. Supplementary Figure 8: Elimination of antigen-specific plasma cells (PC) from the bone marrow using the affinity matrix technology. Bone marrow cells from an OVA-immunized and boosted mouse were collected <15 days after boost, incubated with or without (controls) the matrix, and cultured for 30-45 min (37°C, 5% CO2) in the presence of 10% guinea pig complement. The reaction was stopped, and the cells were stained as described in the methods section. OVA-specific and unspecific cells were enumerated by FACS using a MACSQuant Analyzer. The data are mean ± SE of three replicates of a single experiment, and are representative of two independent experiments. *p ≤ 0.05 by Mann-Whitney U test. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
3
Deletions in immunoglobulin mu chains.
Georges Köhler, Mary Jane Potash, Hans Lehrach et al. · The EMBO Journal · 1982 · 98 citations · Full text