SEARCH

SEARCH BY CITATION

Keywords:

  • B-cell lymphoma;
  • B-cell signalling;
  • Siglecs;
  • SLE

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

CD22, an inhibitory co-receptor of the B-cell receptor, shows a B-cell-specific expression pattern and is expressed on most B-cell lymphomas. The anti-CD22 antibody Epratuzumab is in clinical trials for B-cell non-Hodgkin lymphoma and systemic lupus erythematosus, but shows a mostly unknown mode of action. We generated a new mouse model that expresses human CD22 instead of murine CD22 (Huki CD22 mice), in which human CD22 can be targeted. Expression of human CD22 on the B cells of Huki CD22 mice does not generally interfere with B-cell development. However, Huki CD22 mice show a reduction of the population of mature recirculating B cells in the bone marrow and reduced transitional and marginal zone B cells in the spleen, phenotypes resembling that of CD22-deficient mice. Similarly, enhanced BCR-induced Ca2+ signalling is observed in Huki CD22 mice, which also mount normal immune responses toward different classes of antigens. Huki CD22 B cells show a normal anti-hCD22 antibody-mediated endocytosis. In conclusion, human CD22 cannot fully substitute for murine CD22 functions, possibly due to the changed intracellular tail of the protein or due to lower expression levels. Huki CD22 mice are a valuable new model for both antibody- and immunotoxin-mediated targeting of human CD22.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

Co-stimulating proteins or co-receptors can modulate the B-cell receptor (BCR) signal. One of them is CD22, a member of the Siglec family (sialic acid binding immunoglobulin-like lectins). Siglecs bind to sialic acids, which are usually expressed on cell surfaces and in the extracellular milieu [1, 2]. All Siglecs recognise sialic acids, but they vary in their affinity for different glycoside-linkages. CD22 prefers sialic acids in α2,6-linkage [3-5]. It is an inhibitory co-receptor for BCR-induced calcium (Ca2+) signalling. CD22 carries inhibitory ITIM-signalling motifs, which recruit the tyrosine phosphatase SHP-1 that is responsible for the inhibitory function of CD22. CD22 can bind to α 2,6-linked sialic acid-containing ligands in cis and in trans. This ligand binding can modulate the signalling function of CD22 [3-5].

The murine CD22 and the human CD22 protein share about 60% sequence homology. The human CD22 gene is located on chromosome 19 in the human genome, the mouse gene is located on chromosome 7. The human and the murine CD22 are expressed from the pre-B-cell stage to the mature B-cell stage. CD22 is downregulated on plasma cells [5]. It is also expressed on a majority of all B-cell lymphomas [6], for example, CD22 is expressed on 65% of the B-cell acute lymphoblastic leukaemia [7]. Different immunotoxins use anti-CD22-antibodies to target these lymphoma and leukaemia cells. As a toxin component, these antibodies are linked to bacterial or plant toxins [8], RNases [9], ribosome inactivating proteins [10] or deflycosylating ricin A chains [11]. There is also a humanized monoclonal anti-CD22 antibody, Epratuzumab (LymphoCide®), which was successfully tested in phase II clinical trials for non-Hodgkin lymphoma patients [12].

Dysregulated B cells contribute to the pathogenesis of autoimmune diseases, such as systemic lupus erythematosus (SLE) [13]. CD22 on B cells is used as a target in SLE patients with anti-CD22 antibodies as well. Epratuzumab was clinically tested for SLE patients in phase IIb randomized clinical trials. The studies showed a treatment advantage with Epratuzumab over placebo [14]. The mode of action of Epratuzumab in SLE treatment is not known. Discussed mechanisms are induced ADCC, CD22 internalisation or modulation of B-cell signalling [13].

Our concept for testing and further investigating the mechanism of the immunotoxins and antibodies against human CD22 was to develop a mouse model in which the murine Cd22 gene is replaced by the human CD22 gene. To achieve this goal, we generated a human knockin CD22 (Huki CD22) mouse. These Huki CD22 mice express human CD22 instead of murine CD22 on the surface of their B cells. These mice could be a valuable model for testing the biological effects of anti-human CD22 antibodies and anti-CD22 immunotoxins, which are already in clinical studies both in B-cell lymphoma and SLE patients.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

Generation of the Huki CD22 mouse line

The Huki CD22 mouse line was generated by knockin of the human CD22 cDNA into the mouse Cd22 gene. This results in expression of the human CD22 protein under the control of the murine Cd22 promoter, i.e. under the same transcriptional regulation as the endogenous mouse gene. A targeting vector was generated containing the cDNA of human CD22 in exon 4 of the murine Cd22. By in-frame knockin of the cDNA of human CD22 after the coding sequence for the signal peptide of murine Cd22, we wanted to ensure that the human CD22 will be transported normally to the B-cell surface. E14CreAg Embryonic Stem (ES) cells [15], were transfected with the targeting vector and ES cells were screened by PCR and positive clones confirmed by Southern blot (Supporting Information Fig. 1). Chimeric mice were generated and germline transmission was obtained. Heterozygous and homozygous Huki CD22 mice were analysed for their expression of CD22. Hetero-zygous mice express human and murine CD22 on the surface of B cells of blood, spleen and lymph nodes. Huki CD22 homozygous mice express no murine CD22. On B cells of the spleen, the lymph nodes and the blood, they express human CD22 (Fig. 1A–C). Compared with human peripheral blood B cells stained with the same anti-human CD22 antibody, the expression level of hCD22 on B cells of Huki CD22 mice was lower (about 10–30× lower) (Fig. 1D). Western blot analysis with a polyclonal anti-mCD22 antibody confirmed that Huki CD22 B cells did not express any remaining mCD22 (Fig. 1E).

image

Figure 1. Huki CD22 mice express human CD22 but no murine CD22 on the surface of B cells, as shown by analysis of blood, spleen and lymph nodes. (A) Cells of the blood were stained with anti-B220 versus anti-human CD22 (top) or anti-murine CD22 (bottom). B220+-gated cells are shown. Respective CD22 expression levels of wild-type (WT) mice, heterozygous mice (+/Huki CD22) and homozygous Huki CD22 mice (Huki CD22) are shown. (B) Splenocytes of WT, heterozygous mice (+/Huki CD22) or Huki CD22 mice were stained with anti-B220 versus anti-human CD22 (top) or anti-murine CD22 (bottom). (C) Lymph node cells of WT, heterozygous mice (+/Huki CD22) or Huki CD22 mice were stained with anti-B220 versus anti-human CD22. (D) Comparison of expression levels of hCD22 on B220+ B cells of Huki CD22 mice (grey-shaded histogram) or on CD20+ human peripheral blood B cells (open histogram), both stained with anti-human CD22. (E) Total lysates of B cells of WT, Huki CD22 mice or of human Daudi cells were stained with rabbit anti-mCD22. (A–E) Data are representative of at least three experiments.

Download figure to PowerPoint

Impaired maturation and homing to the bone marrow of Huki CD22-derived B cells

Huki CD22 mice showed a normal B-cell development with no significant changes in pro-, pre- and immature B-cell numbers in the bone marrow (Supporting Information. Table 1). However, the number of recirculating B cells (IgDhigh, B220high) in the bone marrow was reduced (Fig. 2A). In the spleen, the number of transitional B cells (T1 and T2) was reduced, as was the number of marginal zone B cells (MZ). Follicular and mature B cells were found in normal numbers (Fig. 2B). These changes in B-cell populations resemble the phenotype of the CD22 knockout mouse (CD22−/− mouse) (Supporting Information Table 1) [16-20]. These changes of B-cell subpopulations in bone marrow and spleen were not observed in heterozygous (+/Huki CD22) mice (Supporting Information Table 1).

image

Figure 2. CD22−/− and Huki CD22 mice have decreased numbers of mature B cells in the bone marrow and decreased numbers of transitional and MZ B cells in the spleen. Adoptive transfer of B cells shows a decreased migration of Huki CD22 B cells to the bone marrow. (A) Bone marrow recirculating B cells were stained with B220 versus IgD. Representative flow cytometry examples are shown (left). The total number of recirculating B cells (B220+ IgDhi) is shown as mean + SD of eight samples each from eight independent experiments. (B) Splenic cells were stained with IgM versus IgD to determine T1/MZ cells, T2 cells and mature B cells (top) or with CD21 versus CD23 to determine MZ and FO (follicular) B cells (bottom). Representative flow cytometry examples are shown (left). The total number of recirculating B cells (B220+ IgDhi) is shown as mean + SD of five samples each from five independent experiments. (C) Adoptive transfer of splenic B cells. Huki CD22 or WT-derived splenic B cells were labelled with CFSE, i.v. injected into WT mice and stained with B220 versus IgD after 24 h. B220-gated cells of the bone marrow are shown for IgD versus CFSE in a representative flow cytometry plot (left). The relative amount of CFSE-positive cells from IgD-positive cells of spleen and bone marrow is shown as mean + SD of 20 samples each from three independent experiments (right). *p < 0.05, **p < 0.01, Student's t-test. T1: transitional type 1; T2: transitional type 2; MZ: marginal zone; M: mature; FO: follicular B cells.

Download figure to PowerPoint

To address the question whether the decreased number of mature B cells in the bone marrow of Huki CD22 mice is the result of a defect in homing to the bone marrow, we labelled B cells derived from Huki CD22 or WT mice with carboxy-fluorescein succinimidyl ester (CFSE) and injected these cells i.v. into recipient WT mice. This experiment showed that the number of CFSE-labelled cells in the mature B-cell population was higher in the spleen and reduced in the bone marrow, compared with that of WT controls (Fig. 2C). This suggests that the homing/migration or survival of mature Huki CD22 B cells in the bone marrow is indeed impaired, similar to the findings of CD22−/− mice [21].

Higher calcium signial, impaired CD22 phosphoryl-ation and SHP-1 recruitment in Huki CD22 mice

The CD22−/− mice show a higher calcium response after stimulation of the BCR [16-18, 20]. Since the Huki CD22 mice share some B-cell phenotypes with the CD22−/− mice, we analysed the calcium signal of B cells of Huki CD22 mice. After stimulation of the BCR with an anti-IgM F(ab)2, the calcium signal of Huki CD22 B cells was increased, compared with that of control B cells, but did not reach the strength of the signal of CD22−/− B cells (Fig. 3A). In order to determine CD22 activation, we analysed CD22 tyrosine phosphorylation after BCR stimulation. The Huki CD22 B cells or human Daudi cells were stimulated for 0, 2.5, 5 or 10 min with either mouse- or human-specific anti-IgM F(ab)2. While immunoprecipitated hCD22 from Daudi B cells showed an increase of tyrosine phosphorylation and SHP-1 recruitment after BCR stimulation, in B cells of Huki CD22 mice hardly any induced CD22 tyrosine phosphorylation or SHP-1 binding was detected (Fig. 3B). Similarly, when SHP-1 was immunoprecipitated, we could readily detect a phosphorylated 130 kDa band, most likely CD22, in Daudi, but not in Huki CD22 B cells (Fig. 3C).

image

Figure 3. Huki CD22 B cells show enhanced calcium signalling and impaired CD22 tyrosine phosphorylation and SHP-1 recruitment and after BCR stimulation. (A) Calcium mobilisation of Indo-1 loaded splenic B cells of CD22−/−, Huki CD22 and control mice, stimulated with anti-IgM F(ab)2 at the indicated time points. Data shown are representative of results obtained in eight independent experiments. (B) Immunoprecipitation of CD22 and co-immunoprecipitation of SHP-1 in Huki CD22 B cells or human Daudi B cells. Cells were stimulated with either anti-murine or anti-human IgM F(ab)2 at indicated time points and lysed afterwards. CD22 was immunoprecipitated with an anti-hCD22 antibody. After SDS gel electrophoresis and blotting, the membranes were stained with antibodies against phosphotyrosine, SHP-1 and CD22 (on a separately run gel with aliquots of same lysates). Data shown are representative of three experiments. (C) Immunoprecipitation of SHP-1 and co-immunoprecipitation of CD22 in Huki CD22 B cells or human Daudi B cells. Cells were stimulated with either anti-murine or anti-human IgM F(ab)2 at indicated time points. SHP-1 was immunoprecipitated with an anti-SHP-1 antibody. Staining was done with anti-SHP-1 or with anti phospho-tyrosine antibodies. Data shown are representative of two experiments.

Download figure to PowerPoint

Huki CD22 mice show normal serum antibody levels and mount normal antibody responses

Comparing the serum antibody levels of Huki CD22 and WT mice revealed no differences in the IgM, IgG1, IgG2b, IgG3 and IgA levels (Fig. 4A). This is similar to CD22−/− mice that have only mildly increased IgM levels [16-18, 20]. After immunisation of WT and Huki CD22 mice with the thymus independent (TI) type 2 antigen TNP-Ficoll, both groups showed a similar immune response. These results were obtained by a TNP-specific ELISA with anti-IgM and IgG3 specific secondary antibodies (Fig. 4B). The analysis after an immunisation with the thymus-dependent antigen NP-KLH on day 0 and day 21 showed again a similar immune response of WT and of Huki CD22 mice. For this analysis, an NP-specific ELISA with anti-IgM and IgG1 antibodies was performed (Fig. 4C). This means that human CD22 expression does not interfere with B-cell-induced antibody responses.

image

Figure 4. Huki CD22 mice show normal immunoglobulin levels in the sera and have normal thymus-independent type II (TI II) and thymus dependent immune responses. (A) Serum Ig levels of Huki CD22 and WT mice were measured by ELISA. (B) Mice were immunised with TNP-Ficoll at day 0 and anti-TNP IgM and IgG3 responses were determined by ELISA. (C) Mice were immunised with NP-KLH on day 0 and on day 21 and anti-NP IgM and IgG1 responses were determined by ELISA. (A–C) Each symbol represents an individual mouse and the bars represent the means. Data shown are representative of two experiments performed. *p < 0.05, Student's t-test.

Download figure to PowerPoint

Endocytosis of human CD22 on B cells derived from Huki CD22 mice is not impaired

CD22 is a transmembrane protein, which can be internalised efficiently by antibody-mediated endocytosis [22]. To study whether also human CD22 on mouse B cells can be internalised, endo-cytosis of CD22 on B cells after antibody binding was analysed with cells derived from Huki CD22 or WT mice. B cells were isolated and stained with anti-hCD22 or anti-mCD22 monoclonal anti-bodies. After several time points of incubation at 37°C, the antibodies remaining bound to the cell surface were washed away by low-pH buffer and internalised CD22 was analysed after fixation by flow cytometry. Huki CD22 B cells showed a similar efficient CD22 endocytosis (with faster kinetics) as WT B cells expressing murine CD22 (Fig. 5). Due to two different anti-CD22 monoclonal antibodies used, the two types of cells cannot be fully compared.

image

Figure 5. The endocytosis of CD22 on Huki CD22 B cells is not impaired. (A) WT B cells were incubated with anti-murine CD22-PE at 37°C for the indicated times. After this incubation, cells were treated with glycine pH 2.4 and fixed. The MFI was determined; the MFI of the extracellular stain was set as 100%.(B) Huki CD22 B cells were incubated with anti-human CD22-PE at 37°C for the indicated times. The treatment was as in (A). (A, B) Data are shown as mean + SD of three samples and are from one experiment representative of three independent experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

The Huki CD22 mice express human CD22 on the surface of B cells. Murine CD22 could not be detected on these cells, showing that we successfully generated the human CD22 knockin mouse. We then addressed the question whether the human CD22 can replace the murine CD22 functionally. While the Huki CD22 mice showed a normal early B-cell development, some changes in B-cell populations were similar to those of CD22−/− mice. The reduced number of marginal zone B cells is typical for CD22−/− mice [19], as is the reduced number of mature B cells in the bone marrow [21]. Both mouse lines, CD22−/− and Huki CD22 mice, show a reduced number of mature B cells in the bone marrow after adoptive transfer, indicating impaired migration or impaired survival of these cells in the bone marrow. The human CD22 does not seem to be able to replace the murine CD22 in these regards. It was noted, however, that hCD22 was expressed at a lower level in Huki CD22 mice than on primary human blood B cells.

On the other hand, Huki CD22 mice show normal immune response to thymus-independent or thymus-dependent antigens. The antibody serum levels of naïve mice are normal as well. CD22−/− have impaired thymus-independent responses that were attributed to lower MZ B-cell numbers, a population largely responsible for those responses [19]. Despite the reduction of MZ B cells also observed in Huki CD22 mice, in this line the TI-2 response is normal. Apparently, hCD22 can replace mCD22 for these types of responses.

The strength of the calcium signal of Huki CD22 B cells is increased, compared with that of WT B cells, but not to the same extent as in CD22−/− B cells. Immunoprecipitation of human CD22 from Huki CD22 B cells showed impaired tyrosine phosphorylation of CD22 and binding of SHP-1 to CD22 after BCR stimulation, compared with that of Daudi B cells. How strongly hCD22 phosphorylation in Huki CD22 mice is impaired is difficult to quantify, as human CD22 of Huki CD22 B cells cannot be precipitated with the same antibody as mCD22 from control mouse cells. Similarly, human B cells cannot be stimulated in the same way as murine Huki CD22 B cells. From the increased Ca2+ signalling, it could be concluded that human CD22 works only partially as an inhibitory co-receptor on murine B cells, with not as strong inhibitory functions as murine CD22. Why is this the case? A possible explanation is that the cytoplasmic tail of human CD22 cannot perform the same inhibitory functions in mouse B cells as the tail of murine CD22. The three ITIMs of CD22 are crucial for the inhibitory functions [23]. Studies in cell lines suggest that at least two out of three ITIMs are needed for a full inhibitory function of CD22 [24, 25]. Data from CD22 knockin mice generated in our group suggest that all three ITIMs of CD22 are needed. While knockin mice in which all CD22 ITIMs were mutated showed a similar phenotype to the CD22−/− mice, mice with mutations just in the 5th and 6th tyrosines composing two ITIMs, showed a partial phenotype (Obermeier, I., Müller, J., Wöhner, M., Brandl, C. and Nitschke, L., unpublished observations).

Both the cytoplasmic tail of human as well as murine CD22 contain six tyrosines, of which the 2nd, 5th and 6th compose ITIMs [23]. Comparison of the ITIM sequences shows that the sequences around tyrosine 5 and 6 are quite similar in mouse and human (Supporting Information Fig. 2). The sequence around the ITIM of tyrosine 2 is not well conserved. In humans, the amino acid sequence is EDGISYTTLRF. In mice, the sequence is DDTVSYAILRF. Both fit the definition of an ITIM motif (I/L/VxYxxL/V) [26] but the changes in surrounding amino acids might prevent efficient binding of murine SHP-1 to the human ITIM. Changes in other signalling motifs between the human and mouse cytoplasmic tail of CD22 may also contribute to the phenotype of the Huki CD22 mouse. Overall, the sequence identity of the murine and human CD22 tail is 69% (Supporting Information Fig. 2) [27]. Potentially, the changed extracellular domain of CD22 could also contribute to the phenotype of Huki CD22 mice, as the extracellular domain has 62% sequence identity and is less well conserved [27]. The first Ig V-like domain of CD22 binds to 2,6-linked sialic acids. Murine CD22 prefers the Neu5Gc form, while human CD22 binds to both Neu5Gc and Neu5Ac equally well [28]. Thus, potential cis- or trans- ligands interactions should not be affected by the change of the extracellular ligand-binding domain. In addition, we also noted that the human CD22 was expressed at a lower level on mouse B cells in the Huki CD22 mice than on WT primary B cells from human blood. Thus, we cannot exclude that a changed expression level of CD22 may also contribute to the observed phenotype. However, it is unclear whether a similar expression level of human CD22 on mouse B cells as on human B cells is to be expected.

Of note, heterozygous Huki CD22 mice (expressing both human and murine CD22) showed no phenotypic changes, both in numbers of B cells of all B-cell populations and in Ca2+ signalling, when compared with WT mice. Therefore, we have developed a valuable mouse model for targeting human CD22. Antibodies such as Epratuzumab, where the mode of action is not well understood [13]. Experiments with Epratuzumab will be performed in the future. They could not be done yet, because the antibody was not available to us. Alternatively, CD22 immunotoxins can be tested in this mouse model. For immunotoxins, it is of relevance that endocytosis studies in Huki CD22 mice showed that the endocytosis of human CD22 after antibody binding is not impaired. For CD22 targeting studies, the Huki CD22 mouse will be crossed to established transgenic mouse models for both B-cell lymphoma as well as for B-cell-mediated autoimmune diseases.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

Generation of the Huki CD22 mouse line

We generated a targeting vector, containing the cDNA of human CD22, inserted in exon 4 of murine Cd22. As a first cloning step, the polyA signal DNA was cloned into the vector pKStneoloxP from the vector IRES-eGFP (Clonetech). The polyA signal was amplified by PCR, using the primers polyA-Bsu15Ifw (5′-ATATCGATATTCGGACGTGACTGACATATGCGCTCTGTACACCCCCTGAACCTGAAACATAAAATG-3′) and polyA-BamHIrev (5′-ATGGATCCACTCAACCCTATCTCGGTC-3′). The PCR fragment and the targeting vector were digested with the enzymes BamHI and Bsu15I and then ligated. The human cDNA was then isolated from human blood cells using the primers cDNA-PciIrev (5′-TCTGTACATCAATGTTTGAGGATCACATAGTCCAC-3′) and cDNA-Bsu15Ifw (5′-ATATCGATATGCTAAGTAGTGTATTGTAGGACATGTGGCTTCGGCTGACTCAAGTAAATGGGTTTTTGAG-3′). By digesting the cDNA and the vector with the enzymes Bsp1407I and Bsu15I, the cDNA could be cloned into the targeting vector. The long arm of the targeting vector was amplified in two parts. The first part contained the DNA of the murine exons 1 to 3 and part of exon 4 and was obtained from the vector pKSCD22ex1–5R130E (generated in the Nitschke group). By digesting this vector and the targeting vector with the enzymes PscI and Bsu15I, the first part of the long arm could be cloned into the vector. Next followed the second part of the long arm, containing the promoter sequence. A PCR reaction with the genomic DNA of ES cells from the line E14creAG and the primers 22prom-LAfw (5′-TTATCGATGATGCTGTATGGTACTGG-3′) and 22prom-LArev (5′-GCTTTCGCTTGCTGATACATC-3′) was performed. As a last step the short arm, containing the DNA of the murine exon 5, was cloned into the construct. The vector pKS-R130E-Controlplasmid (generated in the Nitschke group) and the targeting vector were digested with the enzymes NotI and SalI and the short arm fragment was cloned into the targeting vector.

The targeting vector already contained loxP sites flanking a neo cassette. The long arm in front on the cDNA now contained the promoter region, exons 1 to 3 and a part of exon 4 of the murine Cd22 gene. The signal peptide, which is responsible for CD22 being transported to the cell surface, is encoded in exon 3 and the first part of exon 4. The human cDNA was cloned in frame into exon 4 of the murine Cd22 gene, 3′ of the sequence coding for the signal peptide.

The targeting vector was opened at the short arm by digesting it with the enzyme NotI and the DNA was then sterilized by ethanol precipitation. The embryonic stem cells from the line E14CreAG (129sv background, expressing protamine-Cre) [15] were transfected with the linearized vector by electroporation. Clones were screened for homologous integration by PCR and verified by Southern blot with an external probe generated with the primers humCD22Sonde3 (5′-CCT CTG TGG GAT TGA CCT TG-3′) and humCD22Sonde4 (5′-CCT CTG ACC ATA GTC ACC TG-3′) and with an internal probe generated by digesting the external probe with the enzyme EcoRI and isolating the 720 bp band. Two positive clones were identified and injected into blastocytes. Those were transferred into pseudo-pregnant females and one gave germline transmission successfully. Mice carrying the correct mutation in their germline were identified with PCR and bred to homozygosity. Age-matched control and Huki CD22 mice of 8–14 weeks were used for analysis.

Flow cytometry

Single cell suspensions of bone marrow, spleen and lymph nodes (cervical, inguinal, mesenteric) were prepared in PBS and 5% FCS. Blood was obtained from tail vein and collected in PBS containing 0.1% heparin (Roche, Basel, Switzerland). After erythrocyte lysis, cells were washed and stained in PBS buffer containing 0.1% (wt/vol) BSA and 0.05% (wt/vol) sodium azide for 30 min at 4°C. The Fc receptors were blocked by saturating concentrations of 2.4G2 (our hybridoma). The following antibodies (conjugated to biotin, FITC, PE, Pe-Cy5, APC) were used: anti-B220 (RA3–6B2, BD Bioscience), anti-IgM (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), anti-CD11b (Mac1) (MI/70, our hybridoma), anti-IgD (11–26C, our hybridoma), anti-CD5 (53–7-3, BD Bioscience), anti-MHC-II-Ab (BD Pharmingen, San Diego, CA, USA), anti-murine CD22 (Cy34.1, BD Pharmingen), anti-human CD22 (eBio4KB128, eBioscience), anti-CD21 (7E9, our hybridoma) and anti-CD23 (B3B4, eBioscience). Biotinylated antibodies were further stained with streptavidin PE-Cy5 or streptavidin PE-Cy5.5. Detection of cell-surface marker expression was performed by flow cytometer (FACSCalibur) and analysed with CellQuest Software (BD Bioscience). Living lymphocytes, judged by forward and side scatter parameters, were gated for analysis. Total cell numbers were obtained by counting after Trypan blue staining of the cells.

Immunisations

Huki CD22 and control mice were immunised intraperitoneally once with TNP-Ficoll (10 μg/mouse) for T-independent responses. Blood samples were taken on days 0, 5, 7, 11 and 14. For T-cell-dependent responses Huki CD22 and control mice were immunised intraperitoneally twice with NP-KLH (100 μg/mouse). The mice were immunised on day 0 and day 21 and blood samples were taken on day 0, 7, 14, 21, 28 and 35.

ELISA

We measured immunoglobulin serum levels from naïve and immunised mice by standard ELISA methods (Gerlach et al. 2003). Maxisorp plates (Nunc) were coated with antigen (10 μg/mL TNP-BSA; 5 μg/mL NP-BSA). As a standard, we used a sample of pooled sera of immunised mice, which was applied to every plate. For the detection of Igs in the sera of naïve mice, we coated polysorb plates (Nunc) with isotype-specific Abs (Southern Biotechnology Associates, Birmingham, AL, USA) and used monoclonal Ig isotype antibodies as standard (Southern Biotechnology Associates). We applied sera as serial dilutions of 1:3 in dilution reagent (1% (wt/vol) BSA, 0.05% (wt/vol) sodium azide in PBS) and incubated plates for 2 h at 37°C. Alkaline-phosphatase-conjugated secondary antibodies were added for detection (Southern Biotechnology Associates).

Calcium measurement

Splenic cells (5 × 106) were resuspended in RPMI 1640 supplemented with 5% FCS and loaded with 4.5 μm Indo-1 plus 0.003 pluronic F-127 (Molecular Probes, Eugene, OR, USA) at 30°C for 25 min and, after adding RPMI with 10% FCS, at 37°C for 10 min. The baseline Ca2+ concentration (proportional to the FL5 to FL4 ratio) was recorded at 37°C using an LSR II (Becton Dickinson, San Jose, CA, USA). Then the cells were stimulated with 13, 6.5 and 3.25 μg/mL anti-IgM (B7.6). B cells were gated as Mac-1 and CD5-negative. Data were analysed using FlowJo software and displayed as overlay histograms of relative fluorescence intensity over time.

Adoptive transfer of CFSE-labelled cells

Splenic cells of donor Huki CD22 or control mice were prepared and a complement-mediated lysis of T cells was performed. The cells were stained with rat IgM anti-mouse CD4, CD8 and Thy1 antibodies on ice. Cells were then washed and incubated at 37°C for 45 min with rabbit baby complement (Cedarlane). Efficiency of T-cell lysis was checked by flow cytometry. Labelling with 0.5 μM carboxyfluorescein succinimidyl ester (CFSE, Molecular Probes, Inc.) was done for 5 min at RT, then cells were washed in supplemented RPMI 1640 containing 10% FCS. The labelled cells were incubated for an additional 30 min at RT in RPMI 1640 containing 10% FCS. Cells were then washed twice with PBS and 1× 107 cells were resuspended in 100 μL PBS and injected intravenously into recipient mice (107 cells per mouse). One day after the B cell transfer, mice were sacrificed. Bone marrow cells and splenic cells were stained with anti-B220 and anti-IgD and analysed for percentage of CFSE-labelled cells by flow cytometry.

Endocytosis of CD22

Huki CD22 and control mice derived B cells (after MACS with CD19 beads) were incubated with anti-human CD22-PE (Santa Cruz Biotechnology) or anti-murine CD22-PE (Cy34.1, BD Pharmingen) for 1 h at 4°C. Cells were then incubated at 37°C, which induced the endocytosis. Samples were taken at 0, 2.5, 5, 10, 30, 60, 120 and 180 min, the samples were washed with 0.2 M glycine/HCl pH 2.5 to remove extracellular bound antibody and then treated with 2% paraformaldehyde for 30 min to fix the cells. After this treatment, only intracellular anti-CD22 antibodies can be detected by FACS analysis. At the time point 0 min a second sample was taken, which was not washed. This sample served as extracellular staining control, the mean fluorescence intensity (mfi) value of this sample was set as 100% and other mfis were calculated in relation to this sample.

Immunoblot analysis

A T-cell depletion was performed with Huki CD22-derived splenocytes (see above). For stimulation, those B cells were incubated with 10 μg/mL F(ab)2 for 2.5, 5 and 10 min at 37°C. Cells were then quickly centrifuged and lysed in 1 mL ice-cold lysis buffer (1 M Tris, pH 7.5, 0,5 mM EDTA, 5M NaCl, 1% Brij 58) with protease and phosphatase inhibitors. For CD22 immunoprecipitation, the lysates were incubated at 4°C overnight with sepharose G and 4 μg anti-human CD22 (mIgG1 monoclonal Ab, Santa Cruz Biotechnology). For SHP-1 immunoprecipitation, the lysates were incubated at 4°C overnight with sepharose A and 5 μg anti-SHP-1 (Millipore). After washing the Sepharose with lysis buffer, the proteins bound to the sepharose were resolved on a 10% SDS-polyacrylamide gel and transferred to a nitrocellulose transfer membrane (Pall). Cell lysated were obtained by lysis of 5 × 106 WT or Huki CD22 splenocytes or Daudi cells with 20-μL ice-cold lysis buffer (1 M Tris, pH 7.5, 0,5 mM EDTA, 5M NaCl, 1% Brij 58). For detection of the proteins, we used the antibodies anti-SHP-1 (Millipore), anti-phosphotyrosine (Millipore) and anti-human CD22 (Santa Cruz Biotechnology). Proteins were visualised by anti-rabbit IgG HRP-conjugated antibody (Cell Signalling) or anti-mouse IgG HRP-conjugated antibody (Bio-Rad) and ECL detection system (Amersham).

Software and statistical analysis

Flow cytometric data were analysed using CellQuest Pro (BD Biosciences) software and FlowJo (TreeStar, Ashland, OR, USA). Statistical analyses were performed using Prism software. For evaluation of significance unpaired t-test was used. Statistical data are presented as mean ± SD.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

We thank Nina Weidner, Anne Urbat, Sieglinde Angermüller, Andrea Schneider and Martina Döhler for technical help. This work was supported by the Deutsche Krebshilfe and the DFG (SFB643).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information
Abbreviations
Huki CD22

human knockin CD22

SLE

systemic lupus erythematosus

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. ACKNOWLEDGEMENTS
  8. Conflict of interest
  9. References
  10. Supporting Information

Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.

FilenameFormatSizeDescription
eji2454-sup-0001-figureS1.pdf190K

Figure S1. Generation of the Huki CD22 mouse line A) Targeting vector scheme. 1-8: exons of murine CD22, neo: neomycin cassette, checked pattern: polyA cassette, arrowheads: loxP sites, crosses indicate the regions of homologous recombination B) Southern Blot of PstI-digested genomic DNA with an external probe outside the short arm. V.6.4 and VI.1.1: names of tested ES-cell clones, arrows indicate the expected bands at 6759 and 7985 bp, dotted line: DNA outside of the targeting vector C) Southern Blot of SacI-digested genomic DNA with an internal probe. V.6.4 and VI.1.1: names of tested ES-cell clones, arrow indicates the expected band at 6600 bp, dotted line: DNA outside of the targeting vector

Figure S2. Amino acid sequence alignement of the cytoplasmatic tail of murine CD22 and human CD22 Gray shows identical sequences, tyrosines are highlighted in fat, grb2 motif is underlined, ITIMs are highlighted by black boxes

Table S1. Absolute cell numers of populations in bone marrow, spleen and peritoneal cavityAbsolute cell numbers for 8- to 12-week old wildtype (WT), CD22-/-, +/Huki CD22 and Huki CD22 mice; values are x 105 for bone marrow and peritoneal cavity and 106 for the spleen; n=6–11 for bone marrow, n=5 for spleen and n=3-5 for peritoneal cavity; *p < 0.05, **p < 0.01, ND=not determined

Please note: Wiley Blackwell 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.