Histochemistry and immunohistochemistry
M. tibialis anterior (TA), M. soleus (Sol) and diaphragm (DIA) muscles were collected from 6 month old male C57BL/10 wild-type mice and mdx mice, respectively. Each group consisted of ≥ 7 mice (control group: n = 7, mdx: n = 8). Left and the right muscles of the hind limb were used. The whole muscle CSA was imaged and analysed in order to obtain less biased results. Altogether, about 2000 fibres were analysed for TA, about 3000 for DIA and 700 for Sol. Muscles were mounted on cork supports using gum tragacanth (Sigma-Aldrich, Steinheim, Germany). The samples were snap-frozen in isopentane cooled with liquid nitrogen and stored at −80°C.
The following procedure involves the use of immunological and histochemical reactions, which result in double staining for fast and slow myosin. Additionally, fluorescence staining of the myofibre membranes allows defining the minimal Feret’s diameter. The staining of myonuclei with the fluorescence dye bisbenzimide H (Hoechst 33258, Sigma-Aldrich) was utilised to identify nuclei located internally within myofibres.
Eight μm serial cross-sections of the muscle of interest were cut at a cryostat temperature of −25°C. We recommend using gelatin coated slides to ensure proper attachment of muscle sections during sequential staining. Slides were cleaned in 70% ethanol (Carl Roth GmbH, Karlsruhe, Germany) for 10 min and rinsed well in de-ionised water for 20 min. 500 ml of a 0.5% gelatin solution (Sigma-Aldrich) containing 0.05% chromium potassium sulphate (Sigma-Aldrich) was prepared at 60°C and filtered through standard filter paper. Slides were incubated in gelatine solution at 37-40°C for about 20 sec. Slides were allowed to dry overnight in a dust free incubator at a maximal temperature of 37°C. Until slides are used, it is recommended to store them at −20°C. Sections can be stored at – 20°C up to a few months.
After one hour at room temperature, sections were covered with 0.1% Triton X-100 (Sigma-Aldrich) in PBS (Phosphate buffered saline; PAA Laboratories GmbH, Pasching, Austria) solution, pH 7.3 for 15 min. Triton-X was removed and the slides were allowed to drain for a few minutes and were then arranged in humid chamber. A 1:100 dilution of slow myosin antibody (Novocastra, Newcastle upon Tyne, UK) in FCS (PAA Laboratories GmbH, Pasching, Austria) was applied to the sections and incubated at 4°C overnight. On the next day, the slow myosin antibody was rinsed off with 0.1% Triton X-100 in PBS and then washed in 0.1% Triton X-100 in PBS for an additional 30 min with one buffer change. Afterwards, the sections were covered with the secondary rabbit anti mouse IgG horseradish peroxidase (DAKO, Glostrup, Denmark) diluted 1:100 in 0.1 M lysine (Sigma-Aldrich), 40% Foetal Calf Serum (FCS) in PBS for 90 min at room temperature. Again, the antibody was rinsed off and sections were washed in 0.1% Triton X-100 in PBS for 30 min with one buffer change refreshing the solution after 15 min. Sections were covered with DAB (3,3’-diaminobenzidine tetrahydrochloride, Sigma-Aldrich) Peroxidase Substrate Solution [5 drops of 1% DAB (20x) added to 5 ml PBS, then mixed with 5 drops of 0.3% H2O2 (20x)] and incubated for 10 min. The DAB (20x) Peroxidase Substrate Solution was prepared as follows. 0.1 g of DAB was added to 10 ml distilled water. 3–5 drops of 10 N HCl (Carl Roth GmbH) were added until the solution turned light brown and mixed for 10 min. Aliquots were stored at −20°C.
Afterwards, sections were rinsed several times in tap water, rinsed briefly in PBS and incubated in FCS for 10 min. Without washing, the second primary antibody against fast myosin (NovoCastra), diluted 1:100 in FCS, was applied and incubated for 60 min. It is important to mention, that we obtained identical results when we interchanged the primary antibodies for slow and fast myosin. Next, the antibody was removed and slides were washed as described before. The sections were incubated with the same secondary antibody mentioned before together with 100 μg/ml Alexa Fluor 488 conjugated WGA (wheat germ agglutinin; Life Technologies GmbH, Darmstadt, Germany). After washing, the sections were covered with Vector SG visualisation solution (Vector Laboratories, Burlingame, CA, USA) for 10 min. 3 drops of chromogen were mixed with 3 drops of hydrogen peroxide substrate. The slides were washed in PBS for 5 min, followed by an incubation in 10 ng/ml Hoechst 33258 (bisbenzimide H) in PBS (50 μl each slide) for 5 min. Afterwards, slides were rinsed in PBS for 2–5 min, followed by a short rinse in distilled water. Nuclear counter stain was performed in Carazzi’s haematoxylin for 30 sec, blued up in running tap water for 2 min and mounted in mounting medium (DakoCytomation Fluorescent Mounting Medium, Dako). We utilised the nuclear counter stain to allow for double-check if nuclei were correctly identified. Carazzi’s hematoxylin solution was prepared as follows. 0.2 g hematoxylin (Merck KGaA, Darmstadt, Germany) was dissolved in distilled water at 50°C. The solution was cooled to room temperature and 0.04 g NaIO3 (Sigma-Aldrich), 10 g AlK(SO4)2 · 12 H2O (Sigma-Aldrich), and 40 ml glycerol (Merck) was added. Images were acquired using a Zeiss Axiovert 200 M fluorescence microscope and a Zeiss AxioCam HR photo camera. Images were taken at 20x magnification.
To register for the analysis service, the user is required to contact S.CO LifeScience GmbH (Munich, Germany) to get access to a personal web portal, which provides the MyoScan analysis module. The user is asked to purchase a volume package for this module type, which defines the number of analyses available. The web based analysis service is accessible worldwide any time. Currently, the costs for the analysis of one specimen are 5–10 €, depending on the prepaid volume package size. Technically, there is no limitation for the file size, apart from general limitations for upload processes via the world wide web. However, if a single file exceeds 20 MB the provider should be informed to ensure proper data handling. All common image file formats including .tif, .jpg, and .gif are supported. The maximum upload file number is limited by the size of the volume package the user has purchased.
Quantitative assessment of muscle histology
Automatic quantitative analysis of muscle histology was performed with a specially designed module MyoScan for the web-based image analysis system S.CORE by S.CO LifeScience GmbH (
http://www.sco-lifescience.com/technology.php5). Three different images of the observed section in the histological slice were taken: 1. Green fluorescent WGA staining for display of the myofibre boundaries, 2. Blue fluorescent Hoechst 33258 staining for display of the nuclei, 3. MHC (myosin heavy chain)-double staining for display of all myofibres which are positive for myosin slow type heavy chain or myosin fast type heavy chain, respectively. In the first image, all membrane structures were separated from background, leading to a mask for the individual muscle myofibres. This mask was superimposed first with the Hoechst 33258 image to identify the nuclei within myofibres resulting in a second mask displaying the relevant nuclei.
Both masks were superimposed with the MHC-double staining image to merge all relevant information in a single image (membrane: white, nuclei: black, MHC-double staining: colored). Based on the Cognition Network Technology of Definiens AG, Munich, relevant structures were extracted from the image as single objects and assigned to different object classes. This allows a detailed analysis of the relevant quantitative and morphometric data, e.g. number of nuclei per fibre and minimal Feret’s diameter.
Finally, each myofibre was automatically annotated with an ID and all data summarised in an Excel file, both as single data for each myofibre and as mean values. The variance coefficient (VC) was determined using the following formula:
Importantly, high image quality is essential to ensure optimal analysis. Therefore, folds, freezing artefacts, uneven or pale staining should be avoided as far as possible. Intensive, high quality staining of the myofibre boundaries is especially important. In our study labeling with WGA worked very well. However, any other membrane or extracellular matrix marker may be used instead. Nevertheless, the user is able to check each individual image processing step generated by MyoScan since result files are provided by the analysis system for individual processing if needed. Thus, the user can identify regions of the sections that are not appropriate for analysis. It is always possible to exclude any inappropriate data from analysis. Moreover, MyoScan applies an algorithm which screens generated images whether identified fibers fit the form and size compared to threshold parameters previously defined by the user.
To verify the data generated by MyoScan, manual analysis was carried out on whole muscle sections of TA muscle (n = 4) of 6 month old
mdx mice using ImageJ software IJ 1.46r [
14]. All sections were analysed for both minimal Feret’s diameter and internally nucleated fibres.