SEARCH

SEARCH BY CITATION

Keywords:

  • dendritic cells;
  • dermatomyositis;
  • inclusion-body myositis;
  • inflammatory myopathy;
  • polymyositis

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Dendritic cells (DCs), immune system cells central to the development of immunity, have not previously been reported in muscle in inclusion-body myositis (IBM). We performed immunohistochemical studies on muscle biopsy specimens from 50 patients using monoclonal antibodies that distinguish two classes of DCs, myeloid DC and plasmacytoid DC. In 17 of 20 IBM and 9 of 10 polymyositis (PM) specimens, myeloid DCs were present in substantial numbers, frequently surrounded and sometimes invading otherwise intact myofibers, and were part of dense collections of cells that included T cells. Dermatomyositis muscle had more plasmacytoid DCs than myeloid DCs, whereas IBM and PM had greater numbers of myeloid DCs. The stellate morphology of myeloid DCs in dense collections of cells that included T cells suggests local intramuscular antigen presentation in IBM and PM. Muscle Nerve, 2006

Dendritic cells (DCs) are central to the development of innate and adaptive immune responses.1 Two main classes of dendritic cells have been distinguished, myeloid DCs and plasmacytoid DCs, and myeloid DCs have been further divided into several subtypes.8 Myeloid DCs are potent antigen-presenting cells that stimulate lymphocytes capable of highly specific immune responses (adaptive immunity), in part through the capture, processing, and presentation of antigen.1 Plasmacytoid DCs play an important role in the innate immune system. They are capable of producing large amounts of interferons alpha and beta (IFN-α/β), which have multiple functions, including stimulating tissues to produce intracellular proteins capable of providing defense against pathogens.2

DCs in muscle tissue in the inflammatory myopathies (IMs), a group of diseases characterized in part by the presence of immune system cells in muscle, have received little study. They were recently identified in muscle in polymyositis (PM) and dermatomyositis (DM),5, 7 and have not previously been reported in inclusion-body myositis (IBM). In DM, we observed abundant plasmacytoid dendritic cells associated with a marked IFN-α/β inducible gene transcriptional response.5 In addition, we demonstrated high expression of the IFN-α/β inducible myxovirus resistance (MxA) protein in capillaries and perifascicular myofibers, characteristic sites of DM pathology. In this study we report the presence and abundance of myeloid DCs in IBM and PM, and differences in subtypes of DC among the inflammatory myopathies.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Patients.

Patients participating in this study and the criteria used for the diagnosis of DM, IBM, and PM have been described previously.4, 5 Briefly, all patients with IBM lacked a family history of muscle disease and had a typical IBM clinical phenotype (preferential involvement of finger and wrist flexors and quadriceps), serum creatine kinase (CK) level of less than 12 times the upper limit of normal, rimmed vacuoles, and inflammatory cells invading nonnecrotic muscle fibers. All other patients with IMs, in addition to satisfying the diagnostic criteria referenced, had objective improvement in manual muscle testing with immunosuppressive therapy. Studies were performed on muscle samples from 50 patients, 45 with IMs and five without a neuromuscular disease. The numbers of patients studied were 20 for IBM, 15 with DM previously reported,5 10 with PM, and five without a neuromuscular disorder. The study was approved by our institutional review boards.

Immunohistochemistry.

Immunohistochemical studies using antibodies directed against the specific plasmacytoid DC marker BDCA-2 were performed as previously described.5 For studies using antibodies directed against the myeloid DC marker BDCA-1, we performed anti–BDCA-1 incubation overnight at room temperature and dilution 1:30, and then sections were incubated with rabbit antimouse immunoglobulins (dilution 1:150, incubation for 15 min; Dako, Carpinteria, California) followed by horseradish peroxidase (HRP) –conjugated polymer bound to goat antirabbit immunoglobulins (rabbit EnVision+ System, incubation for 30 min; Dako). Because of minor reactivity of isotype IgG2a control antibodies, studies were repeated under various conditions, which gave comparable results. These included preliminary treatment with 10% human serum in Dako antibody diluent for 30 min followed by anti–BDCA-1 incubation for 5 h at a dilution of 1:50 and overnight at dilutions of 1:50 and 1:100, followed by incubation with 10% human serum in HRP-conjugated polymer bound to goat antimouse immunoglobulins (mouse EnVision+ System, incubation for 30 min; Dako).

Monoclonal antibodies used were anti–BDCA-1 (Miltenyi Biotech, Auburn, California, clone AD5-8E7, isotype IgG2a) and anti–BDCA-2 (Miltenyi Biotech, clone AC144, isotype IgG1). Additional studies done in selected cases used antibodies to CD3, CD4, CD8, and CD19 as previously described.5 BDCA-1 (also known as CD1c) and BDCA-2 are specific among dendritic cells for myeloid and plasmacytoid DCs, respectively.3 BDCA-1 is known to be expressed by a small population of B cells; CD19 immunohistochemistry was done concurrently to identify such B cells. No other cell type than plasmacytoid DCs has yet been identified as expressing BDCA-2. It is unknown whether all mature tissue plasmacytoid DCs express BDCA-2; plasmacytoid DCs isolated from blood lose BDCA-2 expression upon in vitro stimulation with interleukin-3.3 All studies were done on paired adjacent 10-μm sections to compare BDCA-1 and BDCA-2 immunoreactivity. Controls with mouse monoclonal IgG2a (BD Pharmingen, San Jose, California, clone C1.18.4; Dako, clone DAK-G05, and R&D Systems, Minneapolis, Minnesota, clone 20102.1) isotype matched for the BDCA-1 antibody, IgG1 isotype matched for the BDCA-2 antibody, and Tris-stained sections were done for comparison of nonspecific antibody binding and nonspecific immunoperoxidase reactivity. DAB staining was timed to minimize isotype control staining. Methyl green was used for counterstaining of all sections.

Qualitative and Quantitative Analysis of Dendritic Cell Subtypes.

Qualitative analysis of DC subtypes was performed on all 50 specimens by visual inspection of the overall staining patterns on tissue sections and by comparison of adjacent 10-μm sections, one stained for BDCA-1 and the other for BDCA-2. Quantitative analysis of DC subtypes (Table 1) was performed on muscle samples from 10 patients never treated with immunosuppressive agents (five with DM and five with IBM). The method of cell counting was the same as used in previous quantitative studies of other cell types.5

Table 1. Quantitative analysis of the density of dendritic cells and CD8+ T cells in untreated patients with dermatomyositis compared with inclusion-body myositis.
 Dermatomyositis (n = 5)Inclusion-body myositis (n = 5)
Total (per mm2)PerivascularPerimysialEndomysialTotal (per mm2)PerivascularPerimysialEndomysial
  • Density numbers represent the number of cells per square millimeter of cross-sectional area, of a 10-μm thick section, given as mean ± standard deviation.

  • *

    Data from previously published work.4

BDCA-1 (myeloid DCs)10 ± 64 ± 63 ± 22 ± 224 ± 292 ± 33 ± 518 ± 21
BDCA-2 (plasmacytoid DCs)49 ± 3312 ± 103 ± 233 ± 2814 ± 184 ± 64 ± 56 ± 8
CD8 (cytotoxic T cells)*14 ± 148 ± 91 ± 15 ± 5108 ± 1362 ± 21 ± 1104 ± 137

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

In five nonneuromuscular disease controls, sparse isolated BDCA-1+ (myeloid DCs) and BDCA-2+ (plasmacytoid DCs) cells were seen. Collections of myeloid DCs were present in 17 of 20 IBM specimens and 9 of 10 PM specimens, frequently as widely distributed cells across the section with additional focal high-density accumulations (Fig. 1). Immunoreactivity of the control isotype IgG2a matched for BDCA-1 was variably present to a minor extent, possibly as a consequence of Fc receptors on inflammatory cells present in these infiltrates. The presence of BDCA-1+ cells was inferred by the qualitative difference in staining compared to controls with the same antibody concentrations (Fig. 2). These higher-density accumulations were typically endomysial and either surrounded myofibers (Fig. 3) and sometimes invaded apparently nonnecrotic myofiber regions (Fig. 4), or appeared instead to displace myofibers as dense collections of cells between myofibers (Fig. 5). These dense collections always included CD3+ T cells. In IBM and PM, dense accumulations of DCs surrounding myofibers consisted predominantly of myeloid DCs rather than plasmacytoid DCs (Fig. 6). Quantitative studies were limited to patients with IBM and DM who had not received immunosuppressive therapies. In IBM, the numbers of myeloid and plasmacytoid DCs were similar in perivascular and perimysial locations. However, in endomysial locations myeloid DC numbers were a mean of 3-fold higher than plasmacytoid DCs. In contrast, in DM plasmacytoid DCs (Table 1) were present in numbers 16-fold and 3-fold higher than myeloid DCs in endomysial and perivascular locations. The number of myeloid DCs in IBM muscle was 22% that of CD8+ T cells.4

thumbnail image

Figure 1. Broadly distributed myeloid DCs in IBM. Low-magnification view of diffuse and focal collections of myeloid DCs (black or brown staining). BDCA-1 immunohistochemistry in patient P301 muscle.

Download figure to PowerPoint

thumbnail image

Figure 2. BDCA-1 and control isotype IgG2a staining. Differential staining intensity of (A) BDCA-1 and (B) matched isotype IgG2a control antibodies. Equivalent antibody concentrations of 2 μg/mL with overnight incubations were used in both A and B, adjacent 10-μm sections.

Download figure to PowerPoint

thumbnail image

Figure 3. Myeloid DCs in two patients with IBM. Higher magnification of focal collections of myeloid DC from patients P1 and P38 with IBM. Myeloid DCs surround myofibers (A) and lie in spaces between myofibers (B). Scale bars, 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 4. Myeloid DCs invading nonnecrotic myofiber regions in IBM. Patient 237 with IBM. (A,B) BDCA-1+ myeloid DCs deep within otherwise normal-appearing myofibers. (C,D) Invasion of a myofiber (marked *) by BDCA-1+ myeloid DCs shown in C is compared with a nearby (within 200 μm) section stained with hematoxylin and eosin (H&E) in D to highlight the lack of fiber necrosis present, at least at the indicated distance away. Scale bars, 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 5. Myeloid DCs among dense collections of inflammatory cells. A dense collection of inflammatory cells displaces otherwise intact myofibers. Serial 10-μm sections for (A) CD3, a pan-T cell marker, (B) CD4 a marker of T helper cells, DCs, and macrophages, (C) CD8, a marker for a subpopulation of T cells that are cytotoxic or suppressive, and then 50 μm away for (D) BDCA-1, a marker for myeloid DCs. Note the round or elongated cells in C typical of lymphocytes and the branching dendritic (“stellate”) morphology in D (further highlighted in the inset) typical of activated myeloid DCs. From patient P301 with IBM. Scale bars, 50 μm.

Download figure to PowerPoint

thumbnail image

Figure 6. Myeloid DCs with few BDCA-2+ plasmacytoid dendritic cells in IBM and PM. BDCA-1 and BDCA-2 immunohistochemistry of adjacent 10-μm sections. (A,B) Patient P36 with inclusion-body myositis and (C,D) P80 with polymyositis, myeloid DCs are characterized by cellular staining with BDCA-1 (left panels). There are few or no BDCA-2+ plasmacytoid DCs (right panels).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We report here that myeloid DCs are present in IBM and other IM muscle in substantial numbers, and frequently surround and sometimes invade nonnecrotic myofiber regions. These cells have not previously been identified in IBM muscle tissue, despite their recognition as “the single most central player in all immune responses.”8 The DCs present in muscle in IBM and PM are mostly myeloid DCs, in contrast to DM, which additionally has large numbers of plasmacytoid DCs present.5 The dense collections of DC identified using BDCA-1 in our studies in IBM and PM were not previously noted in one study in PM.7 This previous study used antibody probes reflecting DC maturation status but not specific to myeloid or plasmacytoid lineages. Myofibers were reported as surrounded by DCs, although illustrative figures show one or at most two DCs adjacent to myofibers.

The function of these cells in muscle in IBM and other IMs is uncertain, but two hypotheses are suggested based on their known functions and their arrangement and morphology in IM muscle tissue. First, the presence of myeloid DCs invading nonnecrotic-appearing myofiber regions suggests the occurrence of active phagocytosis, endocytosis, pinocytosis, or receptor-mediated uptake of antigen, activities for which these cells are specialized.1, 8 Myeloid DCs are generally believed to return to lymph node, where they present antigen and activate, or suppress, helper T cells, among other functions. A second hypothesis, however, is that some of these cells are actively presenting antigen and activating T cells locally within muscle tissue, rather than strictly in lymph node. This possibility is suggested by the high density of myeloid DCs with branching, dendritic morphologies within infiltrates containing numerous T cells (Fig. 5). Given the central role of myeloid DCs in immune responses,1, 8 further study of these cells might provide additional insight into the immunological mechanisms occurring in patients with IMs and suggest opportunities for therapeutic intervention based on advances in DC biology.6

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Supported by grants to S.A.G. from the Muscular Dystrophy Association (MDA3523), the Sporadic Inclusion Body Myositis Research Foundation, and the National Institute of Neurological Disorders and Stroke (NIH R01 NS043471) of the National Institutes of Health.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES