Introduction

Sporadic inclusion body myositis (sIBM) is an acquired slowly progressing disorder within the group of idiopathic inflammatory myopathies [1, 2]. Severe progressive muscle weakness accompanied by impairments in physical function are typical clinical traits of the disease [3, 4]. Histologically, sIBM is characterised by muscle fibre damage, inflammatory infiltrates, and over time, marked myofibre atrophy [2, 5, 6].

Studies investigating pathophysiological changes in sIBM muscles have mainly focused on the immune response, including lymphocytes [7, 8], macrophages [9, 10] and major histocompatibility complex [11], while studies describing other histological features such as myofibre size (cross-sectional area: mCSA), myofibre type distribution and the content of myogenic precursor cells are scarce [12,13,14]. Specifically, no previous studies have examined myofibre type-specific (type 1, slow-twitch and type 2, fast-twitch) association of satellite cells (SC), myonuclei, myofibre area per myonuclei (i.e., myonuclear domain: MD) and capillary supply in sIBM skeletal muscle.

Therefore, by analysing fibre-specific myogenic regenerative capacity (SCs and myonuclei), markers of inflammation (M1- and M2 macrophages), and vascularization the present study aimed to expand our understanding of the pathophysiological changes in myocellular morphology related to sIBM.

Methods

Patient characteristics

General characteristics of the included sIBM patients (n = 18) are presented in Table 1. The average age was 68.8 ± 5.9 years and the age at onset of symptoms was 60.1 ± 6.9 years. The patient group included 14 males (78%) and 4 females (22%). Seven patients (39%) were receiving immunosuppressive medication.

Table 1 Baseline characteristics of included patients
Full size table

Study outline

The muscle biopsy material from a previous randomised controlled trial (RCT) conducted in our lab (NCT02317094) was analysed [15]. Only baseline biopsies were included in the present study, comprising muscle biopsies from 21 biopsy-validated sIBM patients. Three biopsies were excluded, as the quantitative analysis was impossible due to excessive tissue damage (Figure A, Supplementary Material) and thus the data from 18 sIBMs muscle biopsies were included in the final analysis. Recruitment procedures have been described in detail previously [15].

Muscle biopsy sampling

Muscle biopsies from m. vastus lateralis (VL) or in cases of severe VL atrophy, m. tibialis anterior (TA) were obtained using the Bergström needle (4.0 mm) technique. Immediately after extraction, the muscle samples were embedded in Tissue Tek (4583, Sakura Finetek, Alphen aan den Rijn, The Netherlands) and frozen in nitrogen-cooled 2-methyl butane. The muscle biopsies were stored in a freezer at − 80 °C until further analysis.

Immunohistochemistry and data processing

Immunohistochemistry procedures were performed as described in detail elsewhere [10]; see Supplementary Table A for a detailed listing of the antibodies used.

The data processing with Visiopharm has been described previously [10]. In brief, the image analysis software (Visiopharm A/S, Denmark) was used following a three-step protocol: (1) Identifying muscle fibres. (2) Identifying and quantification of structures of interest (i.e., satellite cells). The linear distance from any fibre determined the allocation of structures of interest, in terms of fibre type (i.e., closest fibre). (3) Normalising the number of structures of interest to the analysed area (counts/mm2). For type 1 and type 2 fibres, the average analysed area was 1.94 ± 0.19 and 0.74 ± 0.01 mm2 per biopsy sample, respectively. To allow comparisons to the existing literature both density (cells/mm2) and structure of interest/fibre profile” for SCs, macrophages and capillaries are presented in the results section (Tables 3, 4, 5).

Analysis of mCSA, sublaminal myonuclei and centrally-placed myonuclei was conducted by manual analysis (AxioVision SE64 Rel. 4.9.1 software: Carl Zeiss, Germany). First, all cross-sectional fibres were identified, hereafter the cross-sectional area and the relative number of type 1 and 2 fibres, as well as myonuclei and central myonuclei, were recorded. A median of 120 fibres (range [21–837]) were analysed per biopsy (type 1; 71 [10–295] fibres, type 2; 62 [11–542]).

Statistical analysis

All data reported in the present study were analysed for normality and homogenous distribution of variance, which could not be verified uniformly for all outcome parameters. The Wilcoxon’s signed-rank test was used to evaluate differences in outcome parameters between type 1 and 2 myofibres. All descriptive data are presented as group mean ± SD unless otherwise stated. The statistical significance level was set at p ≤ 0.05 (two-tailed testing). Statistical analysis was performed using RStudio (Version 2022.7.1.554) [16].

Results

Myofibre cross-sectional area, myonuclei and myonuclear domain

Mean type 1 fibre mCSA (6040 ± 3054 µm2) was ~ 140% larger than type 2 fibre mCSA (2518 ± 1815 µm2) (p = 0.0002). Histograms depicting fibre type-specific mCSA distribution are shown in Fig. 1.

Fig. 1
figure 1

Relative fibre size distribution in type 1 and type fibres. Blue bars represent type 1 fibres. Orange bars represent type 2

Full size image

Type 1 fibres demonstrated ~ 61% higher myonuclei (Six1+/DAPI) content per fibre profile as compared to type 2 fibres (p < 0.0001). Despite that myonuclear domain was ~ 65% larger in type 1 fibres as compared to type 2 fibres (p = 0.005) (Table 2). The number of centrally placed myonuclei (Six1+/DAPI) was 67% higher in type 2 than in type 1 fibres (p = 0.03) (Table 2); see Fig. 2 for examples of fibre type differences in size (A) and centrally placed myonuclei (B).

Table 2 Myofibre cross-sectional areas (mCSA) and myonuclei
Full size table
Fig. 2
figure 2

Illustrations of muscle biopsy findings from patients with sporadic inclusion body myositis (sIBM) A Type 1 fibres (black) were observed to be larger in size than type 2 fibres (grey). B The number of centrally placed myonuclei (red arrows) was higher in type 2 than in type 1 fibres. C The number of satellite cells (red arrows) was higher in association to type 2 fibres than to type 1 fibres. D M1 (red arrows) and M2 (green arrows) macrophages were observed to group into clusters in the vicinity of small type 2 fibres. A Green = laminin, Grey = MHC fast. B Green = laminin, Grey = MHC fast, blue = DAPI. C Green = laminin, Grey = MHC fast, blue = DAPI, red = pax7. Grey = MHC fast, blue = DAPI, yellow = CD206, red = CD68

Full size image

Satellite cells

Quiescent SC number associated with type 2 fibre area was higher as compared to that of type 1 fibre area (p < 0.0001), and this difference that persisted in the fibre profile analysis (i.e., more SCs/myofibre profile) (Table 3); see Fig. 2C for an example of fibre type differences in quiescent SCs.

Table 3 Satellite cells – cells/mm2 vs cells/fibre profile
Full size table

Proliferating SCs was equally associated with type 1 fibre and type 2 fibre in both the area analysis and the fibre profile analysis (p = 0.08) also when compared with area (p = 0.68; p = 0.08, respectively) (Table 3).

Macrophages

M1 macrophage density in type 2 fibre area was higher as compared to that of type 1 fibre area (p = 0.002). Likewise, M2 macrophages were more densely packed around type 2 fibres (p = 0.01) (Table 4).

Table 4 Macrophages – cells/mm2 vs cells/fibre profile
Full size table

In contrast, in the individual fibre profile analysis, both M1 & M2 macrophage number associated with type 1 were observed to be similar compared to type 2 fibres (p = 0.15; p = 0.34, respectively) (Table 4); see Fig. 2D for an example of fibre type differences in M1 and M2 macrophages.

Capillarisation

Capillary density was of a higher abundance in type 2 fibre area compared to type 1 fibre area (p = 0.0002) (Table 5). However, if related to the fibre profile, type 1 fibres had more capillaries associated with the individual fibre as compared to type 2 fibres (p = 0.008) (Table 5).

Table 5 Capillaries – cells/ mm2 vs cells/fibre profile
Full size table

Discussion

As evaluated for the first time in myositis patients, we observed that mCSA, peripheral myonuclei, myonuclear domain was higher in association to type 1 (slow-twitch) as compared to type 2 (fast-twitch) myofibres. Conversely, quiescent SCs and central myonuclei, were higher in association to type 2 fibres compared to type 1 fibres. For both M1 and M2, the density was observed to be higher in type 2 fibres, however no differences were observed on the “per fibre profile” analyses. No fibre type-specific differences were observed for proliferating SCs. These observations suggest distinct and exclusive pathological changes related to the respective fibre types.

Type 1 hypertrophy and type 2 atrophy

In the present study, mCSA was ~ 140% larger in type 1 fibres as compared to type 2 fibres, which aligns with previous biopsy data (70–100% larger) obtained in sIBM patients [12, 13]. In comparison, healthy age-matched adults (~ 65 years of age) typically demonstrate type 1 fibres that are between 10% smaller and up to ~ 15% larger than type 2 fibres [17,18,19,20]. Our mCSA histogram analysis identified a high number of pathologically large myofibres, sometimes termed megafibres [21], that contributed to the increased mean mCSA, especially for type 1 fibres (cf. Figure 1). Karlsen et al. reported healthy age-matched muscle (VL) to contain less than 10% of type 1 fibres with an mCSA above 7000 µm2 [22], while in the present cohort of sIBM patients ~ 40% of type 1 fibres demonstrated mCSA > 7000 µm2 (cf. Fig. 1).

In healthy lower limb (VL) muscles of adults between 50 and 70 years of age, mCSA of type 2 fibres typically amounts to 4000–6000 µm2, depending on habitual physical activity levels [23]. Specifically, in age-matched muscle biopsies obtained at ~ 65 years of age only a few (< 10%) type 2 fibres demonstrate mCSA ≤ 2000 µm2, irrespective of training history [22]. In contrast, the present group of sIBM patients showed highly marked myofibre atrophy, as 62% of type 2 fibres demonstrated mCSA ≤ 2000  µm2 (cf. Fig. 1 and Suppl. Figure B). These findings underline that the sIBM disease process not only severely affects lower limb muscle fibre morphology in accordance with the previous reports [12, 13], but also supports the findings in a recently published study by Nelke et al., that to our knowledge, is the first to reveal that the atrophic effect of the disease appears to be highly fibre type-specific, preferentially affecting fast-contracting type 2 myofibres [24].

Type 2 fibre atrophy and higher association of quiescent satellite cells

Quiescent SC (Pax7+/Six1+) density associated with type 2 fibres was 52% higher compared to type 1 (cf. Table 3), even though type 1 fibres were larger. In contrast, the number of proliferating SCs (Pax7+/Ki67+) was similarly between type 1 and 2 fibres.

Overall myonuclei number (Six1+/DAPI) was ~ 61% higher in type 1 than type 2 fibres (cf. Table 3), which likely was driven by the larger mCSA of type 1 fibres, given that myonuclei number is known to be positively correlated with fibre size [22, 25, 26]. In contrast, the number of centrally placed myonuclei (Six1+) was ~ 67% higher in type 2 fibres (cf. Table 3), an observation that may be linked to the apparent more substantial type 2 myofiber atrophy.

The present cohort of sIBM patients demonstrated an enlarged (+ 67%) myonuclear domain in type 1 fibres compared to type 2 fibres. This could indicate that the proliferative capacity of satellite cells and their ability to fuse with the type 1 myofibers to donate their myonuclei may be dysregulated, as the mCSA to myonuclei ratio (i.e., myonuclei domain) have been demonstrated to be highly positively associated [27, 28].

Higher degree of inflammation associated with type 2 fibres

Both M1 (CD68 + /CD206-) and M2 (CD68 + /CD206 +) macrophages were found to be threefold more abundant related to type 2 fibres as compared to type 1 fibres. These observations indicate higher levels of proinflammatory activity (M1), and elevated tissue remodelling (M2) associated to type 2 vs. type 1 fibres in sIBM patients. The present study did not obtain fibre type specific data on T-cell content. However, we previously observed spatial overlap between macrophage infiltration and T-cell accumulation in the present sIBM patients, where macrophage infiltration was dominantly concentrated around groups of highly atrophied type 2 fibres [10]. Collectively, the previous and present observations suggest that a fibre type dependent pattern of “disease activity” may exist with sIBM, characterised by vastly atrophied type 2 myofibres, which could be driven by the defective muscle regeneration seen in sIBM and myositis [29, 30].

Metabolically suffering type 1 fibres

The number of capillaries (CD31+) density was greater (~ 65%) in relation to type 2 fibre area than type 1 fibre area. In contrast, observations in healthy age-matched adults typically demonstrate higher capillarisation in type 1 fibres compared to type 2 fibres [20, 31]. This disparity suggests that the sIBM disease might be affecting capillary content, however, based on the present observations it is difficult to conclude whether the disease induces increased type 2 fibre capillarisation, or decreased type 1 fibre capillarisation or both. The lower abundance of capillaries of type 1 fibre area, taken together with the higher myonuclear domain of type 1 fibres (lower abundance of myonuclei per fibre profile), could suggest that type 1 fibres in sIBM are metabolically “suffering”, due to negligence in repair and blood supply. This suggests that the sIBM disease might be affecting angiogenesis, especially within type 1 fibres.

Study limitations

The present data were obtained in a somewhat small and heterogeneous group of sIBM patients, which were characterized by large inter-individual differences in disease duration, plasma creatine kinase levels and immunosuppressive treatment. As the sample size did not allow for more detailed subgroup analysis, cautions should be made against generalisations based on the current findings.” The majority of the muscle biopsies were acquired from m. tibialis anterior, rather than m. vastus lateralis, which could have affected the data, however previous data show no significant difference in inflammation or pathologic features between these muscles [32, 33]. Owing to a high degree of physical inactivity and very pronounced muscle atrophy in some study participants, a number of muscle samples did not allow for quantification of mCSA, SC number, myonuclei content etc. and therefore could not be included in the overall statistical analysis (cf. suppl. Figure A), thus reducing the sample size.

Conclusions

As reported for the first time, sIBM appears to exert differential effects on the morphology as well as density and content of quiescent satellite cells, myonuclei, M1 and M2 macrophages and capillarisation of type 1 and type 2 myofibres. In contrast, no fibre type differences in number of associated proliferating satellite cell could be observed. Somewhat paradoxically, type 2 muscle fibres were characterized by focally elevated levels of quiescent satellite cells and central myonuclei, yet these fibres remained markedly atrophied. Likewise, type 1 fibre morphology was pathologically comprised with the presence of megafibres accompanied by large myonuclear domains. All suggest an impact of the sIBM disease on fibre morphology, presumably due to a preferential imbalance between myofibrillar regeneration and degeneration.