Extranodal lymphoid microstructures in inflamed muscle and disease severity of new-onset juvenile dermatomyositis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 4, April 2009, pp 1160–1172 DOI 10.1002/art.24411 © 2009, American College of Rheumatology Extranodal Lymphoid Microstructures in Inflamed Muscle and Disease Severity of New-Onset Juvenile Dermatomyositis Consuelo M. López de Padilla,1 Abbe N. Vallejo,2 David Lacomis,3 Kelly McNallan,1 and Ann M. Reed1 lar dendritic cells and high endothelial venules. They also expressed high levels of CXCL13 and lymphotoxins known to support lymphoid organogenesis. There were also resident naive CD45RAⴙ T cells and maternally derived B cells and PDCs. Patients with diffuse infiltrates or lymphocytic aggregates were responsive to standard therapy with steroids and methotrexate, but those with follicle-like structures tended to have severe disease that required additional agents such as intravenous Ig or rituximab. Conclusion. These data suggest that lymphoneogenesis is a component of the early disease process in juvenile DM. Ectopic lymphoid structures could indicate a severe course of disease; their early detection could be a tool for disease management. Objective. Juvenile dermatomyositis (DM) is an autoimmune disease of childhood characterized by lesions in skin and muscle that are populated by plasmacytoid dendritic cells (PDCs) and lymphocyte infiltrates. We undertook this study to examine the cellular composition, organization, and molecular milieu of the cellular infiltrates in muscle in juvenile DM and to correlate the infiltrates with clinical disease manifestations. Methods. Since PDCs and lymphocyte foci express CCL19 and CCL21, we investigated for in situ formation of lymphoid microstructures that could be sites of extranodal immune activation. Results. Analyses of muscle biopsy samples from children with new-onset juvenile DM showed 3 categories of lesions: diffuse infiltrates, lymphocytic aggregates lacking follicle-like organization, and follicle-like structures. The last of these exhibited elements of classic lymphoid follicles, including networks of follicu- Juvenile dermatomyositis (DM) is a chronic, multisystem inflammatory disease involving small vessels of skeletal muscle, skin, gastrointestinal tract, and other organs (1). The clinical spectrum is very variable, from mild disease that has minimal functional impact to a chronic, severely disabling condition (1–4). Despite new therapies, juvenile DM remains chronically active in a large proportion of patients. The initiating factors in muscle inflammation are unknown, but immune-mediated processes have profound impact on the pathophysiology of juvenile DM. The cellular infiltrate consists of T cells, B cells, and plasma cells (1,5,6). We have recently reported that many of the CD4⫹ cells in the inflamed muscle of patients with new-onset juvenile DM are CD123⫹ plasmacytoid dendritic cells (PDCs) (7). Since PDCs play a prominent role in adaptive immunity by directly activating naive T cells, their presence in inflammatory lesions in juvenile DM muscle is consistent with an emerging paradigm for a role of PDCs in the pathogenesis of chronic inflammatory disease (8–10). Presented in part at the 71st Annual Scientific Meeting of the American College of Rheumatology, Boston, MA, November 2007. Dr. López de Padilla’s work was supported by the Myositis Association (grant MYOSITIS 1). Dr. Vallejo’s work was supported by the NIH and the Arthritis Foundation. Dr. Reed’s work was supported by the State of Minnesota Partnership, the NIH, the Mayo Foundation, and the Arthritis Foundation. 1 Consuelo M. López de Padilla, MD, Kelly McNallan, BS, Ann M. Reed, MD: Mayo Clinic College of Medicine, Rochester, Minnesota; 2Abbe N. Vallejo, PhD: Children’s Hospital of Pittsburgh, Pittsburgh Cancer Institute, McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania; 3David Lacomis, MD: University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Address correspondence and reprint requests to Ann M. Reed, MD, Division of Rheumatology, Departments of Medicine and Pediatrics, Mayo Clinic, 200 First Street SW, Rochester, MN 55905 (e-mail: firstname.lastname@example.org); or to Abbe N. de Vallejo, PhD, Department of Pediatrics, Children’s Hospital of Pittsburgh Rangos Research Center, University of Pittsburgh, Pittsburgh, PA 15213 (e-mail: email@example.com). Submitted for publication July 14, 2008; accepted in revised form December 30, 2008. 1160 LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS Cellular infiltration in the muscle in new-onset juvenile DM appears to have varying degrees of organization (11). In some patients, PDCs, CD3⫹ T cells, and CD20⫹ B cells are scattered throughout the affected muscle. In other patients, the cells form discrete clusters with discernible T cell and B cell zones. We recently reported (7) that these PDC–lymphocytic aggregates have high levels of CCL19 and CCL21, two chemokines that guide normal lymphocyte traffic inside of secondary lymphoid organs (12). These observations suggest that the inflammatory microenvironment of the muscle in juvenile DM could permit lymphoneogenesis. Lymphoneogenesis is a process whereby cellular infiltrates become organized in a manner similar to that of lymphoid follicles, such that they can even form germinal centers (GCs) within nonlymphoid tissues (13,14). Lymphoid-like microstructures have been reported in some autoimmune conditions (15–18). If such structures are true extranodal lymphoid tissues, then they would express the same molecular effectors as normal secondary lymphoid organs and display similar tissue architecture (12,13,19–23). In this context, we undertook studies to examine the cellular composition, organization, and molecular milieu of the cellular infiltrates in muscle in juvenile DM and correlated the infiltrates with clinical disease manifestations. We searched for networks of follicular dendritic cells (FDCs) and high endothelial venules (HEVs), and because juvenile DM is characterized by microchimerism (24), we examined whether maternallyderived cells are components of lymphoid-like structures in the muscle. PATIENTS AND METHODS Patients and collection of muscle biopsy samples. This study was approved by Institutional Review Boards of the Mayo Clinic and the University of Pittsburgh, with signed informed consent and/or assent. Muscle biopsy samples were obtained from 23 children (ages 2–18 years) with new-onset disease that fulfilled the Bohan and Peter criteria for the diagnosis of juvenile DM (25). Control nonmyositic muscle biopsy samples and waste tonsils were also obtained. Disease activity was assessed using established disease activity tools including muscle enzyme evaluation, manual muscle testing, and the Myositis Disease Activity Assessment Tool and Myositis Damage Index for myositis clinical trials (26). Muscle samples were obtained as part of medical diagnoses. Of the 23 samples, 12 were reported previously (7) and 11 were new. All biopsy samples were examined, and the histologic diagnosis of juvenile DM was confirmed. Control muscle samples (n ⫽ 11) were biopsy samples from patients without myositis, as reported previously (7). In 1161 addition, tonsil tissue was obtained from patients who underwent routine tonsillectomy and was used as control tissue. Tissues were embedded in TissueTek OCT compound (Baxter Diagnostic, McGaw Park, IL) or gum tragacanth (Fisher Scientific, Pittsburgh, PA) and stored at ⫺80°C (7). Frozen sections were cut at 5–8-m thickness and used for hematoxylin and eosin (H&E) staining, immunofluorescence, immunoperoxidase staining, and laser capture microdissection. Histopathologic analysis. Serial cryostat sections from the entire muscle tissue (usually a 1 cm ⫻ 0.5–2–cm sample) were stained with H&E. They were then examined by lowpower field screening to evaluate the patterns of distribution and/or organization of inflammatory cells. Immunofluorescence. Frozen sections were stained by either double or triple immunofluorescence techniques as described previously (7), with fluorochrome-conjugated antibodies to a panel of dendritic cell, T cell, and B cell markers. Staining was done sequentially by combining antibodies to specific cell phenotypes. These included markers for T cells (CD3, CD4, CD8, CD45RA, CD45RO), B cells (CD20, CD23), FDCs (CD21, CNA.42), and HEVs (HECA 452). For a nuclear counterstain, we used 4⬘,6-diamidino-2-phenylindole (DAPI). Slides were analyzed using fluorescence microscopy (Olympus BX51 and Olympus DP70 Microscope Digital Camera; Olympus, Melville, NY) and read independently by at least 2 different observers without access to clinical data. Immunohistochemistry. Expression of CXCL13 (or B cell–attracting chemokine 1 [BCA-1]) and its receptor CXCR5 was examined by double staining according to a standard protocol using biotinylated streptavidin conjugated to horseradish peroxidase (CTS002, CTS003; R&D Systems, Minneapolis, MN). To verify specificity of immunolabeling, primary antibodies were omitted from the staining procedure in a set of control sections (not shown). Quantitative imaging. Whole cross-sections of tissues were systematically examined at high-power magnification. Fluorescent images were selected randomly, focused, captured under a DAPI filter, and recaptured using other filters (fluorescein isothiocyanate autofluorescence and Texas Red) without changing focus to ensure the same plane of focus, and the number of positively staining cells was counted using ImageJ software (NIH Image, National Institutes of Health, Bethesda, MD; online at: http://rsbweb.nih.gov/ij/). The number of positive cells per mm2 was calculated for each high-power field (hpf), and the mean values were calculated. The mean area of a muscle section was 3.0 mm2. Detection of microchimerism in muscle sections by fluorescence in situ hybridization (FISH). We investigated for tissue chimerism on 7-m frozen muscle sections from 6 boys with juvenile DM, by FISH analysis using the CEP X spectrum orange/Y spectrum green DNA probe kit (Vysis, Downers Grove, IL) as previously described (24). Three nonmyositic muscle samples from male subjects were used as controls. To further assess phenotypes of maternal XX cells, the combination of immunoperoxidase staining and FISH was performed on frozen muscle sections as described previously (27). The slides were viewed using both fluorescence and confocal laser microscopy. 1162 LÓPEZ DE PADILLA ET AL Table 1. Demographic and clinical characteristics at disease onset in the children with juvenile dermatomyositis* Pattern Age at onset, mean (range) years Female Cutaneous ulcers Nailfold changes Calcinosis Raynaud’s phenomenon Mechanic’s hands Arthritis/arthralgias GI involvement Fever Avascular necrosis Lymphadenopathy Alopecia Elevated ESR ANA positive Myositis-specific antibodies, anti–Jo-1 Muscle strength, mean† Muscle-associated enzymes, mean‡ Physician’s global assessment of disease activity (0–10-cm VAS), mean Duration of symptoms to time of biopsy, mean (range) months Diffuse (n ⫽ 7) Dense (n ⫽ 11) Lymphoid follicle–like (n ⫽ 5) 16.5 (11–22) 5 (71.4) 0 (0.0) 1 (14.3) 0 (0.0) 0 (0.0) 0 (0.0) 2 (28.6) 0 (0.0) 1 (14.3) 0 (0.0) 0 (0.0) 0 (0.0) 7 (100) 1 (14.3) 0 (0.0) 72 2.4 3.4 7.3 (2–13) 8 (72.7) 1 (9.1) 11 (100) 2 (18.2) 1 (9.1) 2 (18.2) 3 (27.3) 0 (0.0) 2 (18.2) 0 (0.0) 1 (9.1) 3 (27.3) 11 (100) 2 (18.2) 0 (0.0) 70 3.3 4.7 5.6 (4–6) 4 (80) 3 (60) 3 (60) 3 (60) 2 (40) 1 (20) 2 (40) 3 (60) 1 (20) 1 (20) 1 (20) 1 (20) 5 (100) 1 (20) 2 (40) 62 4.8 6.8 5.7 (2–12) 4.8 (1–12) 1.8 (1–3) * Except where indicated otherwise, values are the number (%) of patients. GI ⫽ gastrointestinal; ESR ⫽ erythrocyte sedimentation rate; ANA ⫽ antinuclear antibody; VAS ⫽ visual analog scale. † On the Manual Muscle Test 8, with a highest possible total score of 80. ‡ Mean number of enzymes with abnormal levels (aldolase, creatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase). Detection of microchimerism in peripheral blood. Isolation of CD34⫹ cells. Peripheral blood samples (10 ml) from male patients with juvenile DM were positively separated by CD34⫹ magnetic-activated cell sorting (Miltenyi Biotec, Bergisch Gladbach, Germany). The purity of CD34⫹ blood stem cells was ⬎96% as determined by flow cytometry. FISH of positively selected CD34⫹ cells. Positively selected CD34⫹ blood stem cells were probed for X and Y chromosomes as described previously (24). Only nonoverlapping cells were examined and counted for X and/or Y chromosome–fluorescent signals. Results were expressed as the number of cells per slide. Laser capture microdissection and RNA extraction. Laser capture microdissection from muscle tissue was done using the Veritas Microdissection Instrument (Arcturus, Mountain View, CA) as described previously (7). Laser capture microdissection was carried out only after detailed histologic analysis as described above. Sections used in the histologic studies or serial sections from the same tissue blocks were prepared and then used to capture cellular infiltrates. Real-time quantitative reverse transcription–polymerase chain reaction (RT-PCR) experiments. For relative quantitative analysis, total RNA isolated was reverse-transcribed to complementary DNA (cDNA) using the Reaction Ready kit (SuperArray Bioscience, Frederick, MD). First-strand cDNA was synthesized from 25 ng of total RNA. PCR was performed in duplicate using RT2 real-time SYBR Green/ROX qPCR Master Mix (PA-012; SuperArray Bioscience). The housekeeping gene ␤-actin was amplified to normalize expression levels of target genes between different samples and to monitor assay reproducibility (28). The reaction mixtures without template cDNA were used as negative controls. Melting curve analysis was performed to verify that there was a single PCR product and a lack of primer dimers. For all of the genes measured (i.e., CXCL13, lymphotoxin ␣ [LT␣], LT␤), real-time PCR assays were run with serially diluted standards that were plotted to calculate the exact number of transcripts. Amplification efficiencies of the target and references were equivalent. Statistical analysis. Quantitative data from replicate measurements were expressed as the mean ⫾ SD; data were analyzed using the R statistical package (www.rproject.org). Differences between control muscle samples and the 3 histologic types of juvenile DM muscle were examined by KruskalWallis one-way analysis of variance. Pairwise between-group comparisons (e.g., control versus each histologic type) were examined by Mann-Whitney U test. P values less than 0.05 were considered significant. LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS 1163 Figure 1. Histologic patterns of cellular aggregates in muscle from patients with juvenile dermatomyositis (DM). Serial sections of muscle biopsy samples from patients with juvenile DM and control tonsil tissue were stained with hematoxylin and eosin (H&E) and with antibodies to T cells (CD3⫹; green) and B cells (CD20⫹; red). In one-third of patients examined, T and B cells were distributed as ring-shaped structures resembling lymphoid follicles (A and B) with a B cell–rich center surrounded by T cells (B). Eleven of 23 patients had dense infiltrates of lymphoid cells, but follicle-like structures were absent (C and D). The third group of patients exhibited fewer inflammatory cells, which were diffusely distributed throughout the tissue (E and F). In B, D, and F, arrowheads indicate B cell aggregates; arrows indicate T cells. (Original magnification ⫻ 40.) RESULTS Organization of cellular infiltrates in juvenile DM muscle. All patients had a confirmed diagnosis of juvenile DM (Table 1). Histopathologic evaluation of muscle biopsy samples showed extensive cellular infiltra- tion of perimysial and perivascular areas. Figure 1 shows 3 patterns of organization of infiltrates based on the topographic organization of T and B cells as determined by CD3 and CD20 staining, respectively. These were 1) dense inflammatory infiltrates resembling lymphoid 1164 LÓPEZ DE PADILLA ET AL Figure 2. Architecture of lymphoid follicle–like structures in inflamed muscle from patients with juvenile DM. Muscle tissues with H&E staining patterns of lymphoid follicle–like structures were examined further by multicolor immunofluorescence and immunohistochemistry. A, Tissue sections were stained for the distribution of CD4⫹ T cells (green), CD8⫹ T cells (red), and CD20⫹ B cells (blue). A compact collection of CD4⫹ T cells (arrow) intermingled with CD20⫹ B cells (notched arrowhead) and CD8⫹ T cells (arrowhead) was observed in muscles with follicle-like organization. B, There was also colocalization of CD20 (red) and CD23 (green) with the nuclear stain 4⬘,6-diamidino-2-phenylindole (DAPI; blue) (arrowheads), indicating an advanced differentiation state of B cells. C, Colocalization of CNA.42 (green) and CD21L (red) with DAPI (blue) indicated the presence of follicular dendritic cells (FDCs). Many of the CNA.42⫹CD21L⫹ FDCs (arrow) had fine elongated processes (inset) that formed interdigitating networks between neighboring cells. D, All follicle-like structures were found to contain high endothelial venules (HEVs), identified by staining with HECA 452 (red, in contrast to the blue nuclear stain). Two patterns of HECA 452 staining were observed in juvenile DM muscle: HECA 452 positive (arrow) and HECA 452 negative (arrowhead). (Original magnification ⫻ 100 in A–D; ⫻ 200 in inset.) See Figure 1 for other definitions. follicle–like structures (21.7% of tissues) (Figures 1A and B), 2) dense lymphocytic aggregates lacking folliclelike organization (47.8% of tissues) (Figures 1C and D), and 3) diffuse distribution of inflammatory cells (30.5% of tissues) (Figures 1E and F). The T and B cell compartmentalization in lymphoid-like structures was reminiscent of that seen in tonsil (results not shown). For the majority of muscle sections examined, lymphoid follicle–like structures were located preferentially in the perimysium. None of these patterns was observed in control muscle samples without myositis (not shown). In 2 of 23 examined patients with juvenile DM, interposing follicle-like structures and lymphocytic aggregates were observed. These 2 patients had a higher number of follicle-like microstructures covering the whole area of analyzed tissue. Serial sections from the entire 2-cm open biopsy sample were examined. Muscle-infiltrating lymphoid cells were identified by their reactivity to antibodies to CD3, CD4, CD20, CD23, and CD8 (Figures 1 and 2) as well as by their reactivity to antibodies to CD45RA, CD45RO, CD11c, and CD123. Overall, cellular infiltrates consisted predominantly of CD3⫹ T cells (mean ⫾ SD 59.3 ⫾ 13.41%), CD20⫹ B cells (22.7 ⫾ 14.98%), and CD123⫹CD11c–CD4⫹ PDCs (17.7 ⫾ 3.5%). We previously reported the morphology, location, and distribution of PDCs in inflamed juvenile DM muscle; PDCs were situated adjacent to blood vessels in close proximity to T and B cell areas (7,11). In the present study, follicle-like structures were found in muscle samples from 5 of 23 patients. These structures were characterized by large numbers of CD3⫹ cells distributed as a circumferential ring around B cell and PDC aggregates (Figure 1B). The percentage of B cells and the CD4⫹: LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS 1165 Figure 3. Naive and memory T cells in inflamed muscle from patients with juvenile dermatomyositis (DM). Phenotypes of T cells were examined further by immunofluorescence staining for CD4 (green) (A), the memory marker CD45RO (blue) (B), and the naive marker CD45RA (red) (C). Arrows in A and B indicate CD45RO⫹CD4⫹ T cells. Arrowheads in A and C indicate CD45RA⫹CD4⫹ T cells. There was a predominance of CD45RO⫹CD4⫹ T cells (arrow in D, merged image) surrounded by a small but distinguishable rim of CD45RA⫹CD4⫹ T cells (arrowhead in D, merged image). Naive CD45RA⫹CD4⫹ T cells were found in all juvenile DM muscle with lymphoid follicle–like structures and more occasionally in muscle with lymphocytic aggregates, but not in muscle with diffuse infiltrates. In some cases, naive CD45RA⫹CD4⫹ T cells were found inside blood vessels. (Original magnification ⫻ 100.) CD3⫹ cell ratio were significantly higher in juvenile DM muscle with lymphoid follicle–like structures than in muscle samples having either lymphocytic aggregates or diffuse infiltrates. Muscles with follicle-like structures also had higher proportions of CD8⫹ T cells (Figure 2A). Similarly, cellular infiltrates in juvenile DM muscles with diffuse infiltrates (Figures 1E and F) or lymphocytic aggregates (Figures 1C and D) were predominantly CD3⫹ T cells and CD20⫹ B cells, but there was no discrete organization of the clustering of T cells and B cells. Diffuse infiltrates showed significantly fewer cells, at ⱕ50 cells per hpf; most of these were CD4⫹CD123⫹ PDCs and a few scattered CD20⫹ B cells and CD4⫹ (or CD8⫹) CD3⫹ T cells. Lymphoid cells were generally absent in control muscle (not shown). As expected, tonsil tissues had abundant lymphoid follicles, consisting of CD20⫹ B cells and CD3⫹ T cells found mostly in GCs and in the interfollicular areas, respectively (results not shown). Advanced differentiation state of CD20ⴙ cells in juvenile DM muscle. Since production of autoantibodies to muscle is pathognomonic for juvenile DM (29,30), we examined the differentiation state of the infiltrating B cells. In normal B cells, CD23, the low-affinity IgE receptor, indicates B cell maturation and differentiation to plasma cells (31). Hence, we examined costaining for CD23 and CD20. Results showed high-intensity staining for CD23 in ⬃30% of all CD20⫹ B cells in muscles with lymphoid follicle–like structures (Figure 2B). Costaining 1166 for CD23 and CD20 was observed in only 2 muscle specimens with dense lymphocytic aggregates but lacking follicle-like organization. All of the remaining cases (including muscles with diffuse infiltrates) and all of the control muscle samples were CD23–. FDCs in follicle-like microstructures in juvenile DM muscle. True lymphoid follicles found in secondary lymphoid organs are characterized by networks of FDCs that serve as a reservoir for antigens for intranodal activation of B and T cells (32). Hence, we examined the distribution of FDCs in juvenile DM muscle for the expression of the long form of CD21 (CD21L) and the FDC antigen CNA.42 (33,34). The results showed highintensity staining for these FDC antigens (Figure 2C). CD21L⫹CNA.42⫹ cells showed a dendritic morphology with cytoplasmic projections or elongations. FDCs were found exclusively in juvenile DM muscles with lymphoid follicle–like structures. As expected, tonsil tissue stained strongly for CD21L and CNA.42 (not shown). All control muscle was CD21L–CNA.42–. HEVs in follicle-like microstructures in juvenile DM muscle. HEVs are also considered to be a feature of normal lymphoid follicles (35). We therefore examined patterns of staining for HECA 452, an adhesion molecule expressed on endothelial cells of HEVs (36). In tonsil there was strong HECA 452 staining of blood vessels, as expected (not shown). In juvenile DM muscles, 2 patterns of staining were observed, namely, HECA 452 positive and HECA 452 negative (Figure 2D). Overall, HECA 452–positive vessels were found in 5 of 18 juvenile DM muscle samples, 3 of which had many lymphoid follicle–like structures. HECA 452– positive vessels were observed in 2 juvenile DM muscle tissue samples with dense lymphocyte infiltrates. Juvenile DM muscle with diffuse infiltrates displayed little, if any, HECA 452 staining. HECA 452–positive vessels were absent in control muscle. Naive T cells in follicle-like microstructures in juvenile DM muscle. While naive T cells are found in circulation, they home to, and are residents of, regional lymph nodes (37). Except for ␥/␦ T cells, classic naive CD4⫹ and CD8⫹ T cells are not known to reside in nonlymphoid tissue but have been reported in the peripheral blood of adults with myositis (38). Hence, we further examined the phenotype of T cells in juvenile DM muscles, since the vast majority of CD3⫹ cells were also CD4⫹ (Figure 3). Using antibodies to CD45RO and CD45RA (classic memory and naive markers, respectively), we observed that CD4⫹ T cells within lymphoid follicle–like structures were mostly CD45RO⫹ LÓPEZ DE PADILLA ET AL (Figures 3B and D). In addition, there were significant numbers of CD45RA⫹CD4⫹ T cells within these extranodal follicles (Figures 3C and D). CD45RA⫹ cells constituted 28.7 ⫾ 6.1% of all CD4⫹ T cells in these microstructures. Naive CD45RA⫹CD4⫹ T cells were observed predominantly in perivascular areas and, in some cases, within blood vessels, which suggests that they may have been either infiltrating into, or emigrating out of, the inflammatory lesion (not shown). CD45RA⫹CD4⫹ T cells were also frequently observed in small clusters within CD123⫹ PDC–rich areas (not shown). CD45RA⫹CD4⫹ T cells were sporadically seen in juvenile DM muscle with diffuse infiltrates or with lymphocytic aggregates. These cells were absent in control muscle. Expression of CXCL13, LT␣, and LT␤ in folliclelike microstructures in juvenile DM muscle. The notion of lymphoneogenesis in juvenile DM came from our original observation that CCL19 and CCL21, 2 chemokines involved in the formation of lymphoid follicles, are up-regulated in cellular infiltrates in muscle (7). In the present study, we examined other humoral factors involved in lymphoid organogenesis. We investigated for BCA-1/CXCL13 and CXCR5, a ligand–receptor pair that promotes lymphoid organogenesis through the induction of LT␤ in FDCs (13,15). Lymphoid infiltrates from juvenile DM muscle sections were isolated by laser capture microdissection, and messenger RNA (mRNA) was prepared and used in real-time quantitative RTPCR. While nonmyositic control muscle showed few or no chemokine transcripts, we found that juvenile DM muscle had uniformly high levels of BCA-1/CXCL13 (3.7-fold), LT␣ (2.6-fold), and LT␤ (3.5-fold) transcripts. We found differential levels of expression of these molecules among the 3 histologic types of juvenile DM muscle. Figure 4A shows that diffuse infiltrates had the lowest level of specific mRNA for each of these molecules. Follicle-like structures showed higher levels of up-regulation of CXCL13 and LT␤ transcripts than were found in muscle samples having either lymphocytic aggregates or diffuse infiltrates. Muscles with lymphocytic aggregates showed slightly higher levels of LT␣ mRNA than those with follicle-like organization. To complement the quantitative RT-PCR assays, we examined staining patterns of CXCL13 and CXCR5 in muscle tissue. Figure 4B shows colocalization of these 2 molecules only in follicular structures and lymphocytic aggregates in juvenile DM muscle. As noted above, we observed the highest number of CXCL13⫹ cells within LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS 1167 Figure 4. Expression of CXCL13, lymphotoxin ␣ (LT␣), and LT␤. A, Cellular infiltrates in frozen muscle sections for each of the 3 histologic types were harvested by laser capture microdissection. Messenger RNA was isolated and used in quantitative reverse transcription–polymerase chain reaction (PCR) assays for CXCL13, LT␣, and LT␤. Levels of amplicons for each of the 3 molecules in muscle from patients with juvenile dermatomyositis (JDM) are expressed as the mean fold difference compared with levels in nonmyositic control muscle; data were normalized to the internal housekeeping gene ␤-actin. The amounts of CXCL13, LT␣, and LT␤ were significantly higher in juvenile DM muscle with either dense infiltrates or lymphoid follicle–like structure than in nonmyositic control muscle (P ⬍ 0.05). Juvenile DM muscle with diffuse infiltrates also had higher levels of CXCL13 and LT␣ than did nonmyositic muscle, but the difference was not statistically significant (P ⫽ 0.054). Also shown are agarose gels of fractionated specific PCR products with the indicated amplicon sizes. B, Expression of the receptor-ligand pair, CXCL13 and CXCR5, was verified by immunostaining of frozen sections of juvenile DM muscle with lymphoid follicle–like structures, lymphocytic aggregates, or diffuse infiltrates. Dense staining was observed for CXCL13 (brown; arrows) and CXCR5 (purple; arrowheads) in follicular structures and lymphocytic aggregates. Nonmyositic muscle biopsy samples did not stain for this receptor-ligand pair (not shown). (Original magnification ⫻ 40.) 1168 LÓPEZ DE PADILLA ET AL Figure 5. Microchimerism of lymphoid follicle–like structures in muscle and in peripheral blood from patients with juvenile dermatomyositis (DM). Muscle tissue from male juvenile DM patients with follicle-like structures (A–D) or lymphocytic aggregates (E and F) and peripheral blood CD34⫹ cells (G) were subjected to fluorescence in situ hybridization (FISH). FISH probes for X and Y chromosomes were labeled with spectrum orange (red) and fluorescein isothiocyanate (green), respectively. The merged image in D shows a female XX cell (2 red spots in a single nucleus with intact borders; arrowhead) surrounded by male XY cells (arrow indicates Y chromosome). Shown in E and F is the colocalization of CD20 and the X-chromosome probe. Similar colocalization of the X-chromosome probe and the plasmacytoid dendritic cell marker CD123 was also found (not shown). Arrowheads in B, E, and F show female XX cells. Depicted in G are CD34⫹ cells from the blood of a male juvenile DM patient showing the presence of a female XX cell (2 red spots in a single nucleus with intact borders; arrowhead) along with male XY cells (1 green and 1 red signal; arrow). Nuclei werre counterstained with 4⬘,6-diamidino-2-phenylindole (blue). Immunophenotypes of XX cells were better discriminated in lymphocytic aggregates due to lower cell density and less overlap of cells relative to follicle-like structure. (Original magnification ⫻ 250 in A–D; ⫻ 100 in E–G.) follicle-like structures (Figure 4A). Neither CXCL13 nor CXCR5 was detected in nonmyositic control muscle. Microchimerism. Investigators in our group have reported that juvenile DM is characterized by microchimerism, in which maternal cells can be identified in the inflammatory lesions in the muscle as well as in circulating leukocytes (39). Microchimerism has been postulated to be one of the driving forces of autoimmunity in juvenile DM (39,40). Among the 23 children studied, we analyzed 6 boys. Using FISH, maternal cells, identified by 2 X chromosomes, were detected in muscle biopsy samples from these boys. We observed maternal XX cells in the center of lymphoid follicle–like structures (Figures 5A–D) and throughout the diffuse infiltrates. The phenotype of the maternal XX cells was determined using FISH for X and Y chromosomes simultaneously with immunohistochemical staining for CD20, CD45, and CD123 (Figures 5E and F). The results showed that 33% of maternal XX cells expressed CD20 and 25% expressed CD123. None of the cells expressed CD45. The rest were uncharacterized female cells. Chimerism was further examined in blood from 3 boys with juvenile DM by FISH analysis of sorted CD34⫹ blood cells (Figure 5G). The total number of XX cells in the blood was 1–10 cells per 2,500 CD34⫹ cells with distinct X and Y chromosomes (0.04–0.42%) (data not shown). These data suggest that XX cells found in blood from boys with juvenile DM come from a hematopoietic precursor that is maternally derived. Whether maternal XX stem cells are themselves found in muscle or are derived from muscle stem cells is unknown at this time, since muscle tissues were not available for muscle stem cell analysis. Clinical features associated with the prevalence of follicle-like microstructures in juvenile DM muscle. The 17 female patients and 6 male patients with juvenile DM all manifested characteristic skin rash, symmetric proximal weakness, elevated levels of muscle-derived enzymes, electromyographic evidence of myopathy, and classic perifascicular atrophy with capillary occlusion on LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS muscle biopsy. Table 1 shows demographic and clinical characteristics and laboratory findings in all patients with juvenile DM, distributed according to the histologic pattern of cellular infiltrates in the muscle. For the purpose of our study, severe disease was defined as muscle strength ⬍68/80 by Manual Muscle Test 8 (MMT 8) scoring (26), a score of ⱖ5 on a 10-cm visual analog scale (VAS) for physician’s and patient’s global assessment of disease activity, and evidence of cutaneous and muscle vasculopathy with cutaneous ulcerations. The disease activity was scored at diagnosis and after 6 months of followup. We found that patients with lymphocytic aggregates were more likely to have severe disease, but that they responded to standard therapy with oral and intravenous corticosteroids and methotrexate. In contrast, patients with well-defined lymphoid follicle–like structures had severe disease and still needed additional disease-remitting agents such as cyclosporine, intravenous Ig, or rituximab after 6 months of standard treatment. Patients with juvenile DM who were diagnosed as having milder disease (defined as an MMT 8 score of ⱖ70, no cutaneous ulcerations or periungual telangiectasias, physician’s and patient’s global assessment VAS scores of ⬍2.5, and responsiveness to treatment) all had diffuse infiltrates. These observations suggest that outcome and response to treatment are correlated with the presence of ectopic lymphoid structures in muscle lesions. DISCUSSION Lymphocytic aggregates are characteristic of inflammatory myopathies in both adults and children (1,6,41). Our data show that some of these cellular aggregates organize as extranodal lymphofollicular structures in the muscle of children with newly diagnosed juvenile DM. Dense aggregates of CD3⫹ T cells and CD20⫹ B cells were found in almost all juvenile DM muscle biopsy samples examined. Approximately one-third of all biopsy samples had the following cellular and molecular components reminiscent of lymph nodes: 1) abundant lymphoid infiltrates with discrete compartmentalization of CD3⫹ T cells and CD20⫹ B cells, 2) accumulation of naive CD45RA⫹CD4⫹ T cells, 3) HEVs, 4) activated CD23⫹ B cells, 5) CD8⫹ T cells, 6) FDC networks, and 7) the lymphoid-specific chemokines CXCL13, LT␣, and LT␤. Complementing our previous report of high levels of expression of CCL19 and CCL21 (7), the present data categorically indicate 1169 formation of ectopic lymphoid microstructures. The muscle inflammatory process is focal and can at times spare closely associated tissue or have disparate effects on muscles in close proximity. In the diagnosis of juvenile DM, biopsy samples from multiple muscles are generally not clinically obtained. In the present study, all specimens were open biopsy samples with an average yield of a 2-cm3 sample, minimizing sampling error. With this single random biopsy approach, it is noteworthy that we were able to identify discrete histologic forms of lymphoid organization in juvenile DM muscle that even correlated with clinical disease severity. Lymphocytic aggregates with characteristics of extranodal lymphoid structures in inflammatory lesions have been reported for 2 other autoimmune diseases, Sjögren’s syndrome (inflamed salivary glands) and rheumatoid arthritis (the synovium) (15,42). In both of these diseases of adults, the follicle-like structures in the lesions are populated with T cells, B cells, and FDCs similar to what we report here for children with newonset juvenile DM (43,44). The presence of dendritic cells in extranodal follicular structures in these 2 adult diseases has not been validated. In contrast, our present data show that PDCs and FDCs are components of lymphoid follicle–like structures in juvenile DM. Although the presence of classic GCs has yet to be determined, our data indicate that lymphoid-like microstructures in juvenile DM muscle likely represent extranodal sites of immune activation. They are populated by activated PDCs in close proximity to T cells and B cells. Accumulation of naive CD45RA⫹CD4⫹ T cells strongly suggests bypassing of immune activation and perhaps even immune regulation that normally occur in regional lymph nodes. While the overexpression of CXCL13 suggests active homing of naive and memory T cells to the inflammatory lesions, it is not yet clear whether antigen-driven activation and expansion of T cells occur in situ. Investigators in our group previously demonstrated maternal chimerism in juvenile DM (24,39) and other autoimmune disorders. In maternal chimerism, akin to graft-versus-host disease, it is thought that maternally derived T and B cells could be reactive with the host (the child with juvenile DM) (45). An alternative view is that chimerism could be beneficial to the host. In this case, maternally derived cells could actively induce tolerance in the host. In the present study, we found maternal XX cells in the muscle of boys with juvenile DM, particularly in B cell–rich areas, and in follicle-like structures in general. 1170 Our data show that these maternal cells express antigens that tag B cells (CD20) and PDCs (CD123, CD4). We also identified CD34⫹ XX cells in the blood of boys with juvenile DM, indicating that maternal hematopoietic stem cells are present in the blood in juvenile DM. It is not known whether XX cells found in the juvenile DM muscle lesion could also be derived from maternal muscle stem cells. Perhaps the most compelling evidence for lymphoneogenesis in juvenile DM is provided by our observations of the high levels of expression of CXCL13, LT␣, and LT␤, as well as the presence of HEVs; these findings complement our previously reported findings on the expression of CCL19 and CCL21 (7). CXCL13, CCL19, and CCL21 are well-known chemokines for homing of lymphocytes and dendritic cells to normal lymphoid organs (46). LT␣ and LT␤ are critical factors for lymphoid organogenesis, so much so that targeted deletion of the encoding genes in mice yields animals with a tiny amount of or completely absent lymphoid tissue (22). Further, notwithstanding the role of HEVs in intranodal trafficking and peripheral export of lymphocytes, HEVs are essential for lymphoid organogenesis since they form the conduit through which precursor cells of FDCs and stromal cells enter the developing lymphoid organ (46). HEV staining was not seen in all the ectopic lymphoid structures in juvenile DM muscle. The reason for this is unknown, but it could be related to angiogenic processes of HEV formation, which remain to be examined in the context of lymphoneogenesis. Nevertheless, high levels of expression of CXCL13, CCL19, CCL21, and HEVs in juvenile DM muscle suggest a local microenvironment that is permissive for extranodal lymphoid development in new-onset disease. It could be that development of these structures is among the initiating events that ultimately lead to muscle inflammation. An important finding of the present study is the association of lymphoid follicle–like structures with severe disease. In clinical practice, evaluation of muscle biopsy in the diagnosis of juvenile DM has become less common. Our histologic and molecular data, in conjunction with clinical data on muscle strength and vascular changes, underscore the importance of tissue histopathology in ascertaining juvenile DM prognosis (3,47). We report the unique observation that muscle lymphoid follicle–like organization is found among patients with more severe and difficult-to-treat disease, while those with milder disease have a scattering of inflammatory cells. Large numbers of naive and memory T cells, B cells, FDCs, PDCs, and HEVs in these follicle-like LÓPEZ DE PADILLA ET AL structures indicate that they are extranodal foci of immune activation that bypass the normal regulatory controls in the regional lymph nodes. It could be that these structures represent inflammatory foci that are more difficult to disrupt; hence our observation that patients having these structures required additional therapeutic agents. Thus, it will be of interest to examine whether indications of clinical improvement in difficultto-treat juvenile DM with anti–B cell therapy using rituximab (48,49) might involve disruption of the folliclelike structures via the in situ depletion of B cells. Having identified the cellular and molecular components of these extranodal structures, we suggest that they are potential targets of improved immune-based therapies especially for juvenile DM. ACKNOWLEDGMENTS We thank Molly S. Wagner for assistance with chimerism analysis; Anthony Blahnik and Jim Tarara for assistance with laser capture microdissection and confocal microscopy, respectively; Dr. Andrew Engel (Mayo Clinic Muscle Laboratory) for providing additional specimens; and José R. Padilla for assistance with the creation of the figures. AUTHOR CONTRIBUTIONS Dr. Reed had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. López de Padilla, Vallejo, Reed. Acquisition of data. López de Padilla, Lacomis, McNallan, Reed. Analysis and interpretation of data. López de Padilla, Vallejo, Lacomis, McNallan, Reed. Manuscript preparation. López de Padilla, Vallejo, Lacomis, Reed. Statistical analysis. López de Padilla. REFERENCES 1. Banker BQ, Victor M. Dermatomyositis (systemic angiopathy) of childhood. Medicine (Baltimore) 1966;45:261–89. 2. Pachman LM. Juvenile dermatomyositis: immunogenetics, pathophysiology, and disease expression. Rheum Dis Clin North Am 2002;28:579–602. 3. Miles L, Bove KE, Lovell D, Wargula JC, Bukulmez H, Shao M, et al. Predictability of the clinical course of juvenile dermatomyositis based on initial muscle biopsy: a retrospective study of 72 patients. Arthritis Rheum 2007;57:1183–91. 4. Ramanan AV, Feldman BM. Clinical features and outcomes of juvenile dermatomyositis and other childhood onset myositis syndromes. Rheum Dis Clin North Am 2002:28:833–57. 5. Dalakas MC. Immunopathogenesis of inflammatory myopathies. Ann Neurol 1995:37 Suppl 1:S74–86. 6. De Bleecker JL, Engel AG, Butcher EC. Peripheral lymphoid LYMPHONEOGENESIS IN JUVENILE DERMATOMYOSITIS 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. tissue-like adhesion molecule expression in nodular infiltrates in inflammatory myopathies. Neuromuscul Disord 1996:6:255–60. Lopez de Padilla CM, Vallejo AN, McNallan KT, Vehe R, Smith SA, Dietz AB, et al. Plasmacytoid dendritic cells in inflamed muscle of patients with juvenile dermatomyositis. Arthritis Rheum 2007;56:1658–68. Cavanagh LL, Boyce A, Smith L, Padmanabha J, Filgueira L, Pietschmann P, et al. Rheumatoid arthritis synovium contains plasmacytoid dendritic cells. Arthritis Res Ther 2005;7:R230–40. Jahnsen FL, Farkas L, Lund-Johansen F, Brandtzaeg P. Involvement of plasmacytoid dendritic cells in human diseases. Hum Immunol 2002;63:1201–5. Cella M, Jarrossay D, Facchetti F, Alebardi O, Nakajima H, Lanzavecchia A, et al. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat Med 1999;5:919–23. Lopez de Padilla CM, Reed AM. Involvement of dendritic cells in autoimmune diseases in children. Pediatr Rheumatol Online J 2007;5:16. Luther SA, Bidgol A, Hargreaves DC, Schmidt A, Xu Y, Paniyadi J, et al. Differing activities of homeostatic chemokines CCL19, CCL21, and CXCL12 in lymphocyte and dendritic cell recruitment and lymphoid neogenesis. J Immunol 2002;169:424–33. Hjelmstrom P. Lymphoid neogenesis: de novo formation of lymphoid tissue in chronic inflammation through expression of homing chemokines. J Leukoc Biol 2001;69:331–9. Weyand C, Kurtin P, Goronzy J. Ectopic lymphoid organogenesis: a fast track for autoimmunity. Am J Pathol 2001;159:789–93. Takemura S, Klimiuk PA, Braun A, Goronzy JJ, Weyand CM. Lymphoid neogenesis in rheumatoid synovitis. J Immunol 2001; 167:1072–80. Page G, Lebecque S, Miossec P. Anatomic localization of immature and mature dendritic cells in an ectopic lymphoid organ: correlation with selective chemokine expression in rheumatoid synovium. J Immunol 2002;168:5333–41. Amft N, Curnow SJ, Scheel-Toellner D, Devadas A, Oates J, Crocker J, et al. Ectopic expression of the B cell–attracting chemokine BCA-1 (CXCL13) on endothelial cells and within lymphoid follicles contributes to the establishment of germinal center–like structures in Sjögren’s syndrome. Arthritis Rheum 2001;44:2633–41. Armengol MP, Juan M, Lucas-Martin A, Fernandez-Figueras MT, Jaraquemada D, Gallart T, et al. Thyroid autoimmune disease: demonstration of thyroid antigen-specific B cells and recombination-activating gene expression in chemokine-containing active intrathyroidal germinal centers. Am J Pathol 2001;159:861–73. Weninger W, Carlsen HS, Goodarzi M, Moazed F, Crowley MA, Baekkevold ES, et al. Naive T cell recruitment to nonlymphoid tissues: a role for endothelium-expressed CC chemokine ligand 21 in autoimmune disease and lymphoid neogenesis. J Immunol 2003;170:4638–48. Aust G, Sittig D, Becherer L, Anderegg U, Schutz A, Lamesch P, et al. The role of CXCR5 and its ligand CXCL13 in the compartmentalization of lymphocytes in thyroids affected by autoimmune thyroid diseases. Eur J Endocrinol 2004;150:225–34. Drayton DL, Ying X, Lee J, Lesslauer W, Ruddle NH. Ectopic LT␣␤ directs lymphoid organ neogenesis with concomitant expression of peripheral node addressin and a HEV-restricted sulfotransferase. J Exp Med 2003;197:1153–63. Drayton DL, Liao S, Mounzer RH, Ruddle NH. Lymphoid organ development: from ontogeny to neogenesis. Nat Immunol 2006;7: 344–53. Aloisi F, Pujol-Borrel R. Lymphoid neogenesis in chronic inflammatory diseases. Nat Rev Immunol 2006;6:205–17. Reed AM, McNallan K, Wettstein P, Vehe R, Ober C. Does 1171 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. HLA-dependent chimerism underlie the pathogenesis of juvenile dermatomyositis? J Immunol 2004;172:5041–6. Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). N Engl J Med 1975;292:344–7. Oddis CV, Rider LG, Reed AM, Ruperto N, Brunner HI, Koneru B, et al, for the International Myositis Assessment and Clinical Studies Group. International consensus guidelines for trials of therapies in the idiopathic inflammatory myopathies. Arthritis Rheum 2005;52:2607–15. McNallan KT, Aponte C, el-Azhary R, Mason T, Nelson AM, Paat JJ, et al. Immunophenotyping of chimeric cells in localized scleroderma. Rheumatology (Oxford) 2007;46:398–402. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2⫺⌬⌬CT method. Methods 2001;25:402–8. Love LA, Leff RL, Fraser DD, Targoff IN, Dalakas M, Plotz PH, et al. A new approach to the classification of idiopathic inflammatory myopathy: myositis-specific autoantibodies define useful homogeneous patient groups. Medicine (Baltimore) 1991;70:360–74. Wedderburn LR, McHugh NJ, Chinoy H, Cooper RG, Salway F, Ollier WE, et al, on behalf of the Juvenile Dermatomyositis Research Group (JDRG). HLA class II haplotype and autoantibody associations in children with juvenile dermatomyositis and juvenile dermatomyositis–scleroderma overlap. Rheumatology (Oxford) 2007;46:1786–91. Delespesse G, Sarfati M, Wu CY, Fournier S, Letellier M. The low-affinity receptor for IgE. Immunol Rev 1992;125:77–97. Van Nierop K, de Groot C. Human follicular dendritic cells: function, origin and development. Semin Immunol 2002;14:251–7. Liu YJ, de Bouteiller O, Parham CL, Grouard G, Djossou O, de Saint-Vis B, et al. Follicular dendritic cells specifically express the long CR2/CD21 isoform. J Exp Med 1997;185:165–70. Raymond I, Al Saati T, Tkaczuk J, Chittal S, Delsol G. CNA.42, a new monoclonal antibody directed against a fixative-resistant antigen of follicular dendritic reticulum cells. Am J Pathol 1997; 151:1577–85. Girard JP, Springer TA. High endothelial venules (HEVs): specialized endothelium for lymphocyte migration. Immunol Today 1995;16:449–57. Duijvestijn AM, Horst E, Pals ST, Rouse BN, Steere AC, Picker LJ, et al. High endothelial differentiation in human lymphoid and inflammatory tissues defined by monoclonal antibody HECA-452. Am J Pathol 1988;130:147–55. Lee Y, Chin RK, Christiansen P, Sun Y, Tumanov AV, Wang J, et al. Recruitment and activation of naive T cells in the islets by lymphotoxin ␤ receptor-dependent tertiary lymphoid structure. Immunity 2006;25:499–509. O’Hanlon TP, Messersmith WA, Dalakas MC, Plotz PH, Miller FW. ␥␦ T cell receptor gene expression by muscle-infiltrating lymphocytes in the idiopathic inflammatory myopathies. Clin Exp Immunol 1995;100:519–28. Reed AM, Picornell YJ, Harwood A, Kredich DW. Chimerism in children with juvenile dermatomyositis. Lancet 2000;356:2156–7. Artlett CM, Ramos R, Jiminez SA, Patterson K, Miller FW, Rider LG. Childhood Myositis Heterogeneity Collaborative Group. Chimeric cells of maternal origin in juvenile idiopathic inflammatory myopathies. Lancet 2000;356:2155–6. Arahata K, Engel AG. Monoclonal antibody analysis of mononuclear cells in myopathies. I. Quantitation of subsets according to diagnosis and sites of accumulation and demonstration and counts of muscle fibers invaded by T cells. Ann Neurol 1984;16:193–208. Stott DI, Hiepe F, Hummel M, Steinhauser G, Berek C. Antigendriven clonal proliferation of B cells within the target tissue of an autoimmune disease: the salivary glands of patients with Sjögren’s syndrome. J Clin Invest 1998;102:938–46. Nordmark G, Alm GV, Ronnblom L. Mechanisms of disease: 1172 44. 45. 46. 47. primary Sjögren’s syndrome and the type I interferon system. Nat Clin Pract Rheumatol 2006;2:262–9. Sarkar S, Fox DA. Dendritic cells in rheumatoid arthritis. Front Biosci 2005;10:656–65. Kremer Hovinga IC, Koopmans M, Baelde HJ, de Heer E, Bruijn JA, Bajema IM. Tissue chimerism in systemic lupus erythematosus is related to injury. Ann Rheum Dis 2007;66:1568–73. Muller G, Lipp M. Concerted action of the chemokine and lymphotoxin system in secondary lymphoid-organ development. Curr Opin Immunol 2003;15:217–24. Wedderburn LR, Varsani H, Li CK, Newton KR, Amato AA, LÓPEZ DE PADILLA ET AL Banwell B, et al, on behalf of the UK Juvenile Dermatomyositis Research Group. International consensus on a proposed score system for muscle biopsy evaluation in patients with juvenile dermatomyositis: a tool for potential use in clinical trials. Arthritis Rheum 2007;57:1192–201. 48. Noss EH, Hausner-Sypek DL, Weinblatt ME. Rituximab as therapy for refractory polymyositis and dermatomyositis. J Rheumatol 2006;33:1021–6. 49. Cooper MA, Willingham DL, Brown DE, French AR, Shih FF, White AJ. Rituximab for the treatment of juvenile dermatomyositis: a report of four pediatric patients. Arthritis Rheum 2007;56:3107–11.