Human and mouse specimen Preparation
Tissue samples were fixed in a 10% buffered formalin solution, embedded in paraffin and sliced into flat Sect. (10-µm-thick sections were prepared for SAM, whereas 4-µm-thick sections were prepared for LM). Enzymatic digestion makes the surface irregular and causes the section to detach from the glass slide. To protect the detachment, we used an immune-coated slide for immunostaining (Muto Pure Chemicals, Tokyo, Japan).
For mouse bone tissue, the bones were provided by Dr. Y. Enomoto from the Department of Regenerative & Infectious Pathology and fixed and soaked in a 0.5 mol/L ethylenediaminetetraacetic acid (EDTA) solution (Fujifilm Wako chemicals, Tokyo, Japan) for 2 days for decalcification.
For the cytology section, residual free cells from ascites or pleural effusions were prepared to make single-cell-layer slides using a previously reported liquid-based cytology method (BD CytoRich™; Franklin Lakes, NJ, USA)12. The slides were fixed in 95% ethanol for 10 min and postfixed in 10% buffered formalin for 45 min.
Enzyme digestion
To digest the sections, various enzymes were used, including actinase E (pronase E) (Funakoshi, Tokyo, Japan), collagenase type 2 (Worthington, Lakewood, NJ, USA), DNase 1 (Merck, Darmstadt, Germany) and α-amylase (Fujifilm Wako, Osaka, Japan). Actinase and collagenase type 2 were dissolved at 1 mg/mL in phosphate-buffered saline (PBS) containing 0.5 mM CaCl2. α-Amylase (10 mg/mL) was dissolved in PBS (pH 7.4), and DNase 1 (0.1 mg/mL) was dissolved in 20 mM HCl containing 1 mM MgCl2 and 1 mM CaCl2. The enzyme solution was mounted on the section and incubated at 37 °C. The activity of each enzyme determined the incubation duration. The sections were washed in distilled water at each time point and observed with SAM. After observation, the same section was reincubated in the enzyme solution.
SAM observations
We used a SAM system (AMS-50AI; Honda Electronics, Toyohashi, Aichi, Japan) with a central frequency of 320 MHz and a lateral resolution of 3.8 μm, as previously reported24,25,26. The tissue or cytology section was placed on the stage, and distilled water was used as the coupling fluid between the transducer and the section. The waveforms reflected from the surface and bottom of the sample were compared to measure the AOS at each point8. The waveform from the glass surface without a specimen was considered the zero AOS area (black) and was used as the reference. The transducer scanned the sections along the X- and Y-axis for a few minutes to capture the images. The plotted AOS value at each point generated an AOS image.
LM observation
LM slides taken from near the SAM section locations or the same section as for the SAM observation were prepared for comparison with the corresponding AOS images. Staining methods, including haematoxylin and eosin, Elastica Masson trichrome, Congo red and periodic acid-Schiff (PAS), were the same as the routine histology methods employed in the pathology laboratory.
Immunohistochemistry
We utilised the Dako REAL EnVision detection system using the peroxidase reaction with DAB for immunohistochemistry and followed the analysis procedure. The primary antibody was anti-Ki67 (MIB-1, DAKO). For antigen retrieval, histological and cytological sections were soaked in 10 mM Tris-EDTA buffer (pH 9.0) (Abcam, Tokyo, Japan) at 95 °C for 40 min. After immunostaining, the sections were counterstained with haematoxylin.
Transmission electron microscopic observation of paraffin sections
TEM observation of formalin-fixed paraffin sections was performed using previously reported methods27,28. DAB-stained sections were fixed with 2% glutaraldehyde for 60 min for pre-fixation and then incubated with 2% osmium tetroxide for 15 min for post-fixation. The sections were dehydrated with an alcohol gradient and embedded in epoxy resin (Quetol 812, Nisshin EM, Tokyo, Japan) by heating at 60 °C for 48 h. Ultrathin 70-nm-thick sections were prepared, stained with lead and uranium acetate (Merck), and observed by JEM 1400 Plus (JEOL, Tokyo, Japan).
Statistical analysis
The means and standard deviations (SD) of the AOS values were calculated from at least five areas per slide. Mean AOS values between different tissue components and at different time points following protease digestion were compared using Student’s t-test or Welch’s t-test, as appropriate. A commercial statistical software package (BellCurve for Excel; Social Survey Research Information, Tokyo, Japan) was used to calculate the mean values for areas-of-interest, generate dot plot graphs and perform t-test analyses. Before statistical analysis, all continuous datasets were tested for normal distribution. A p-value of < 0.05 was considered statistically significant.