SEARCH

SEARCH BY CITATION

Keywords:

  • autoimmunity;
  • autoreactive B cells;
  • plasmacytes;
  • SLE

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Targeting autoreactive B lymphocytes at any stage of their differentiation could yield viable therapeutic strategies for treating autoimmunity. All currently used drugs, including the most recently introduced biological agents, lack target specificity. Selective silencing of double-stranded DNA-specific B cells in animals with spontaneous lupus has been achieved previously by the administration of a chimeric antibody molecule that cross-links their DNA-reactive B cell immunoglobulin receptors with inhibitory FcγIIb (CD32) receptors. However, long-lived plasmacytes are resistant to this chimeric antibody as well as to all conventional treatments. Bortezomib (a proteasome inhibitor) depletes most plasma cells and has been shown recently to suppress disease activity in lupus mice. We hypothesized that the co-administration of non-toxic doses of bortezomib, that partially purge long-lived plasma cells, together with an agent that selectively silences DNA-specific B cells, should have additive effects in an autoantibody-mediated disease. Indeed, our data show that the simultaneous treatment of lupus-prone MRL/lpr mice with suboptimal doses of bortezomib plus the chimeric antibody resulted in the prevention or the delayed appearance of the disease manifestations as well as in a prolonged survival. The effect of the combination therapy was significantly stronger than that of the respective monotherapies and was comparable to that observed after cyclophosphamide administration.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

All drugs used at present to treat patients with autoimmune diseases lack specificity. This applies to the agents that have been in use for decades, such as cyclophosphamide (CY) and methotrexate, as well as for the novel biological agents including anti-CD20 and anti-Blys monoclonal antibodies [1-5]. They all affect various disease-irrelevant cells and tissues, which results in frequent and serious immediate as well as long-term side effects. In addition, some patients fail to respond to any available treatment. Previous efforts to target and silence selectively cells involved in the autoimmune process, such as double-stranded (ds) DNA-specific B lymphocytes in lupus, were only partially successful [6, 7].

The analysis of V-genes of immunoglobulin (Ig)G anti-self antibodies has shown that autoreactive responses are antigen-driven [8-10]. Thus, one could speculate that all autoreactive (including dsDNA-specific) B cells could be controlled by the same mechanisms that regulate the intensity and the duration of antibody responses to foreign antigens. One of these mechanisms is the cross-linking of the antigen-specific B cell receptors (BCR) with inhibitory CD32 (FcγIIb) receptors by IgG immune complexes. Tchorbanov et al. have made a successful attempt to silence dsDNA-specific B cells selectively by a chimeric antibody that cross-links CD32 (FcγIIb) with dsDNA-binding BCRs. However, this artificial molecule was immunogenic in mice, as it was constructed using a rat monoclonal antibody backbone [11]. Immunogenicity is a problem even when antibodies of mouse origin are used to construct similar chimeric antibodies [12].

A promising new group of therapeutic agents are the proteasome inhibitors. Bortezomib (Bz), a selective inhibitor of the 26S proteasome, has been approved for the treatment of refractory multiple myeloma and mantle cell lymphoma. When proteasome activity is inhibited defective immunoglobulin chains accumulate, causing overwhelming endoplasmic reticulum stress and terminal unfolded protein response, resulting in apoptotic cell death [13, 14]. The same mechanisms also make normal plasma cells, including the long-lived ones, sensitive to proteasome inhibition. The administration of Bz has been shown to reduce autoantibody levels significantly and to prolong survival in Murphy Roths large (MRL)/lpr and New Zealand black (NZB)/New Zealand white (NZW) lupus mice [13]. Proteasome inhibitors are toxic agents, especially after long-term use. In addition, they act non-selectively on all plasma cells regardless of the specificity of the antibodies produced.

We hypothesized that the co-administration of a low dose of a proteasome inhibitor with an agent silencing targeted autoreactive B lymphocytes should have important beneficial effects on the course of an autoantibody-mediated autoimmune disease because: (i) dsDNA-specific B lymphocytes will be silenced selectively and the entry of new cells with the same specificity into the pool of mature plasma cells will be suppressed; (ii) the population of pre-existing long-lived plasma cells, which are resistant to all current treatments, will be reduced; and (iii) the undesirable antibody response to the chimeric molecule will be delayed or even prevented. In the present study we demonstrate that the repeated co-administration to lupus-prone MRL/lpr mice of the chimeric molecule together with suboptimal doses of Bz strongly suppresses disease activity and prolongs survival, an effect that was comparable to the one seen after CY treatment.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Pharmaceutical preparations

Two licensed drugs were used in the experiments: cyclophosphamide (Endoxan®, from Baxter Oncology GmbH, Weiterstadt, Germany) and bortezomib (Velcade®, from Janssen-Pharmaceutica NV, Beerse, Belgium). Stock solutions of both were prepared and kept frozen at −80°C.

Chimeric antibody molecule

The antibody chimera selectively targeting dsDNA-specific mouse B lymphocytes was a gift from Dr Andrey Tchorbanov (Institute of Microbiology, Sofia, Bulgaria). The chimera has been constructed by coupling copies of a dsDNA-mimicking peptide to the rat anti-mouse CD32 (FcγIIb) receptor monoclonal antibody 2·4G2, as described previously [11].

Experimental animals

Female MRL/lpr mice were purchased from Charles River Laboratories (Saint Germain sur l'Arbresle, France) and were kept in specific pathogen-free conditions. All manipulations were approved by the Animal Care Commission of the Institute of Microbiology.

Animal experiments

Groups of 10 female 7-week-old MRL/lpr mice were treated as follows: (i) weekly with 1 ml phosphate-buffered saline (PBS) pH 7·2 intraperitoneally (i.p.) (control group); (ii) weekly with 25 mg/kg CY i.p.; (iii) once every 2 weeks with 0·75 mg/kg Bz i.v. (the dose of Bz used in the present experiments was lower than in a previous study [13]); (iv) twice weekly with 20 μg/mouse of the antibody chimera intravenously (i.v.); and (v) with Bz plus the chimeric molecule at the same doses and time schedules described above.

Blood samples were collected twice monthly and the sera were stored at −20°C. Treatment with the chimeric antibody was ceased at the age of 3 months because of the appearance of antibodies to it. All other treatments continued until the death of the mice.

Physical examination of the animals

All mice were examined weekly by an independent observer for the development of the characteristic skin lesions and of lymphadenopathy. The degree of the latter was scaled from 0 to 3, as follows: 0, no enlargement of lymph nodes; 1, an increase of at least one lymph node on one side that is not visible but is felt by palpation; 2, visible two-sided lymph node enlargement which does not affect the mobility of the mouse; and 3, extremely enlarged lymph nodes so that the animal's mobility is hindered.

Proteinuria measurement

The proteinuria levels were measured weekly using urine test strips (Combi Screen, Analyticon Biotechnologies AG, Lichtenfels, Germany). The results were presented semiquantitatively: 0, no proteinuria; 1, 30 mg/dl; 1·5, 30–100 mg/dl; 2, 100 mg/dl; 3, 100–500 mg/dl; and 4, >500 mg/dl.

Enzyme-linked immunosorbent assay (ELISA)

The ELISA assay for IgG anti-dsDNA antibodies was performed as described previously [11]. Antibodies to the chimeric molecule and its components were also determined by ELISA. Sera from the mice treated with the chimeric molecule plus Bz were diluted 1:100 and incubated for 1 h in the wells of plates coated with: (i) the dsDNA-mimicking peptide [15] conjugated to bovine serum albumin (Sigma-Aldrich, St Louis, MO, USA); (ii) the rat monoclonal 2·4G2 antibody; or (iii) the chimeric antibody (all at 10 μg/ml). The following steps were performed as described in Tchorbanov et al. [11].

Pathohistological examinations

Kidneys from 18-week-old MRL/lpr mice were frozen and the cryostatic sections were stained with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG antibody (Sigma-Aldrich). The glomerular deposition of immune complexes was observed using a fluorescence microscope (Carl Zeiss, Oberkochen, Germany). Parts from the same kidneys were fixed in formalin and incorporated in paraffin blocks, cut into sections and stained with haematoxylin and eosin. The mononuclear infiltration around the glomeruli, kidney tubuli, the blood vessels and interstitium was rated as: 0, absent; 1, low-grade; 2, moderate; and 3, strong.

Statistics

The significance of the differences in IgG anti-dsDNA antibody levels, proteinuria and lymphadenopathy grade was evaluated by Student's t-test. Analysis of variance (anova) with repeated measurements could not be used because of missing data. To reduce the trauma and blood loss, different mice were bled on different days, which produced an incomplete data set. The Kruskal–Wallis non-parametric anova test with multiple comparisons as post-estimation was used for evaluation of the degrees of skin and renal pathology. Survival curves were compared by the log-rank test. Data analysis was performed using the GraphPad Prism 4 software package (GraphPad Software Inc., La Jolla, CA, USA) (for all methods *P < 0·05; **P < 0·01; ***P < 0·001).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Simultaneous administration of the antibody chimera plus Bz delayed anti-dsDNA antibody production and proteinuria more strongly than the monotherapies

The present study confirms the previously shown ability of Bz to decrease the plasmacyte population in both 18- and 24-week-old lupus-prone animals [13]. The mice from the PBS-treated group in this study had measurable IgG anti-dsDNA antibody and proteinuria levels at 18 weeks. As expected, CY strongly delayed their appearance, whereas the effects of the chimeric antibody or of the suboptimal doses of Bz injected alone delayed the onset of the antibody production by 3 and 5 weeks, respectively. Their co-administration, however, prevented the increase of anti-dsDNA IgG antibodies and of proteinuria. The latter effect was significantly more potent than that of two agents injected alone and was comparable to the effect of CY (Figs 1 and 2).

figure

Figure 1. Simultaneous administration of the proteasome inhibitor plus the chimeric molecule resulted in a significant reduction of anti-dsDNA IgG autoantibody levels. The horizontal grey arrows show the duration of the treatments. Differences in the levels of the anti-dsDNA IgG autoantibodies between the phosphate-buffered saline (PBS)-injected and all other groups were significant (*P < 0·05; **P < 0·01; ***P < 0·001, Student's t-test). The differences between the combination therapy group and the groups treated with bortezomib or the chimeric antibody were also significant (not shown).

Download figure to PowerPoint

figure

Figure 2. Proteinuria levels in the lupus-prone mice treated with the agents under study. The horizontal grey arrows show the duration of the treatments. Differences in proteinuria levels between the phosphate-buffered saline (PBS)-injected and all other groups were significant (*P < 0·05; **P < 0·01; ***P < 0·001, Student's t-test). The differences between the combination therapy group and those treated with bortezomib or the chimeric antibody were also significant (not shown).

Download figure to PowerPoint

Co-administration of the antibody chimera plus Bz prevented lymphadenopathy and skin lesions

Disease progression in the Fas-deficient MRL/lpr mice is accompanied by progressing lymphadenopathy due to defective apoptosis and the resultant massive accumulation of T and B lymphocytes in lymphoid organs. PBS-injected animals developed severe enlargement of the cervical, auxiliary and inguinal lymph nodes. Interestingly, lymphadenopathy was reduced significantly after the administration of CY, Bz or the antibody chimera as well as in the group treated simultaneously with the chimera plus the proteasome inhibitor. The effect of Bz, injected alone, was equal to that after the combination therapy, whereas monotherapy with the chimeric molecule was less effective (Fig. 3). All administered treatments also suppressed strongly the appearance of severe inflammatory skin lesions seen in the control animals (Fig. 4).

figure

Figure 3. The monotherapies with cyclophosphamide (CY), bortezomib (Bz) and the antibody chimera as well as the co-administration of Bz plus the chimera prevented the lymphadenopathy in MRL/lpr mice. The arrows correspond to the duration of the respective treatments. The differences between the measured scores in the phosphate-buffered saline (PBS) and the other groups are significant (*P < 0·05; **P < 0·01; ***P < 0·001, Student's t-test). Lymphadenopathy was significantly lower in the mice subjected to the combination therapy when compared to those that received the chimeric molecule only, but not to the animals treated with Bz (not shown).

Download figure to PowerPoint

figure

Figure 4. All treatments suppressed the development of the inflammatory skin lesions in the lupus-prone mice. Representative photographs and percentages of animals with cutaneous involvement in the individual studied groups (on the right, *P < 0·05; **P < 0·01, Kruskal–Wallis test).

Download figure to PowerPoint

Combination therapy prevented glomerular IgG immune complex deposition and modified renal injury

Lupus glomerulonephritis and the resulting renal failure are responsible for the short life-expectancy of the MRL/lpr mice. Immunofluorescence studies showed no IgG immune complex depositions in the glomeruli of Bz- or Bz plus chimeric antibody-treated animals. The effect of the chimeric molecule administered alone was weaker, and the treatment with CY had no effect (Fig. 5, upper panels). No significant differences were seen in the mononuclear infiltration scores in the glomeruli of the mice of all experimental groups. Administration of the chimeric antibody and combination therapy diminished the mononuclear infiltration significantly around the blood vessels, kidney tubuli and in the interstitium. The effect of the combination therapy was equal to that of the chimeric antibody injected alone (Fig. 5, lower panels).

figure

Figure 5. The administration of bortezomib alone or in combination with the antibody chimera prevented the glomerular deposition of immunoglobulin (Ig)G-immune complexes (left panels – the white arrows point to glomeruli). The middle panels display representative haematoxylin and eosin (H&E)-stained kidney sections from the same kidneys. The mononuclear infiltration of different kidney compartments in animals from the studied groups was scored from 0 to 4 (right panels, **P < 0·01; ***P < 0·001, Kruskal–Wallis test).

Download figure to PowerPoint

Combination treatment prolonged the survival of the lupus-prone MRL/lpr mice

Monotherapies with the low dose of the proteasome inhibitor or with the chimera alone had no effect on survival. However, their co-administration prolonged the lifespan of the animals significantly. The effect was similar to that observed after CY treatment (Fig. 6).

figure

Figure 6. Survival curves of the treated MRL/lpr mice. The simultaneous administration of the proteasome inhibitor plus the antibody chimera prolonged the life-span of the animals significantly (*P < 0·05, log-rank test). The effect was comparable to that of cyclophosphamide (CY).

Download figure to PowerPoint

Administration of the proteasome inhibitor failed to suppress the antibody response to the chimeric antibody

The simultaneous administration of the proteasome inhibitor plus the chimeric molecule failed to prevent the immune response to the chimeric molecule. Antibodies to the intact chimera as well as to its monoclonal IgG backbone and the dsDNA-mimicking peptide were detected after the sixth week after the start of the treatment (not shown). The dose of Bz was obviously too low to prevent the induction of chimera-neutralizing antibodies.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

The current study confirmed the hypothesis that the co-administration of two agents targeting, respectively, the pathological autoreactive dsDNA-specific B lymphocytes and plasma cells can have additive effects in suppressing lupus activity in MRL/lpr mice. When started in young, disease-free animals, this treatment delayed the appearance of some manifestations of the disease (anti-dsDNA IgG antibodies, proteinuria) and prevented the glomerular deposition of immune complexes, lymphadenopathy and severe inflammatory skin lesions. Monotherapy with a low, suboptimal dose of Bz or the antibody chimera had only a short-lived suppressive effect on anti-dsDNA IgG antibodies and on proteinuria levels. In a previous study the chimeric antibody treatment has been shown to suppress specifically dsDNA-specific B cells and their differentiation into IgG anti-dsDNA antibody-producing plasma cells. This treatment improved survival of MRL/lpr mice only when its administration was started in animals with full-blown disease, but not at the age of 7 weeks [11]. In the latter case the disease could re-establish itself after the end of the 8-week-long course of chimeric molecule injections. Survival of the young, disease-free animals was, however, improved significantly in the present experiments in which the same treatment was combined with the administration of a plasma cell-purging drug. The silencing of dsDNA-specific B cells, resulting in decreased cell numbers that enter into the plasma cell pool [11] and the simultaneous partial elimination of already present plasma cells by low doses of the Bz, successfully delay the development of this aggressive autoimmune disease.

The severe cutaneous inflammatory lesions and lymphadenopathy in the studied Fas-deficient mouse strain are known to be due to a massive accumulation of T cells [16-21]. The chimeric antibody and Bz target B lineage cells; however, Bz is also known to reduce double-negative, double-positive and CD4+CD25+ T cells in the thymus [22]. The chimeric antibody itself may also have an indirect effect on T cells [23]. IgG-containing immune complexes were observed in the glomeruli of CY- and chimera-injected, but not of Bz- and Bz plus chimera-treated 18-week-old animals. Interestingly, the administration of the proteasome inhibitor prevented IgG immune complex deposition, but not the development of renal disease. The lack of correlation between immune complex deposition and the severity of the renal pathology comes as no surprise, as severe lupus glomerulonephritis could develop even in the complete absence of circulating antibodies, stressing the potential role of T cells as well as the antibody-independent role of the B lymphocytes in the pathogenesis of the disease [24].

Signalling through the inhibitory CD32 (FcγIIb) could be amplified further if its B cell surface expression is enhanced. This has been achieved by the exposure of splenocytes from non-autoimmune Balb/c as well as from MRL/lpr mice to pooled therapeutic human IgG (IVIg). IVIg pretreatment enhanced significantly the ability of the chimeric antibody to suppress the differentiation of dsDNA-specific B lymphocytes to IgG antibody-producing plasma cells [25, 26]. The treatment strategy proposed here could be improved further if a tri-specific chimeric antibody that targets both CD32 (FcγIIb) and CD22 inhibitory receptors on dsDNA-specific B lymphocytes is used, as signalling through the first receptor silences B cells with IgG antigen receptors; and through the second, those with IgM BCRs [12, 27].

Chimeric antibody molecules constructed by protein engineering techniques are immunogenic. We hoped to solve this problem by co-administering the modified antibody with a plasma cell eliminating drug. However, twice-monthly Bz administration failed to prevent the immune response of lupus-prone mice to the antibody chimera. Resolving the problem of immunogenicity requires novel approaches [28]. For the last few decades CY has been the drug of choice for the treatment of lupus patients with renal involvement. We used this as a standard [2] to which the effects of experimental therapies could be compared. The present study proves that CY could be replaced by a more selective therapeutic approach.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

The authors are grateful to Dr Anastas Pashov for careful reading of the text. This work was supported by grants from the Bulgarian National Science Fund (TK-X-1710/07), the Swiss National Science Foundation (IB73B0-110719) and DFG, FOR 832 (VO 673/3-1).

Disclosures

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References

Authors declare no financial or commercial conflicts of interest.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosures
  9. References