We performed miRNA microarray experiments using frozen tissues from nine MCL lymph node specimens collected from the Department of Pathology of Kurume University (Fukuoka, Japan) (Table I). MCL was diagnosed according to the World Health Organization (WHO). MCL samples analyzed in this study, all of which were included in authors' previous study (10), are confirmed to carry the chromosomal translocation t(11;14)(q13;q32) (IgH/CCND1) and expression of CCND1 and SOX11 proteins. Categorization of MCL in this study were done based on WHO classification. The details are declared in the previous study (9). Briefly, ‘Classical type’ has morphological characteristics of classical MCL in WHO classification. ‘Aggressive type’ includes morphological features of blastic variant or pleomorphic variant of MCL in WHO classification. ‘Intermediate type’ has morphological transition in one section from region classified in classical type to region done in aggressive type. Of all 9 MCL specimens, 4 specimens were categorized in classical type (cMCL), 4 specimens were aggressive type (aMCL), and 1 specimen was intermediate type (iMCL, n=1). For iMCL, the region of classical MCL (iMCL-c, n=1) and the region of aggressive MCL (iMCL-a, n=1) were obtained by laser microdissection (LMD). This study was approved by the Kurume University Institutional Review Board and in accordance with the Declaration of Helsinki. Informed consent was not obtained because the data were analyzed anonymously.

The tissue samples were immediately frozen in acetone/dry ice and stored at −80°C for microdissection. The iMCL sample was embedded in an optical cutting temperature (OCT) compound (Sakura Finetek, Tokyo, Japan) and frozen in liquid nitrogen. Cryosections (20 μm-thick) were mounted on 2.0 μm-thick PEN-Membrane slides (MicroDissect GmbH, Herborn, Germany). After fixation in 100% ethanol, the slides were stained rapidly with toluidine blue O (Chroma-Gesellschaft Schmid GmbH & Co., Köngen, Germany), washed with diethylpyrocarbonate (DEPC)-treated water, and air-dried using a fan.

The frozen sections were microdissected with a Leica LMD6000 laser microdissection system following the manufacturer's protocol (Leica, Wetzlar, Germany). The classical or aggressive parts were microdissected from the same iMCL tissue sections with LMD. The dissected cells were collected in 0.5-ml tube caps filled with 50 μl lysis buffer for RNA extraction (10,16).

Total RNA was isolated from frozen whole tumor tissue samples (cMCL and aMCL) with TRIzol (Invitrogen, Carlsbad, CA, USA). For the laser-dissected samples (iMCL-c and iMCL-a), total RNA was extracted with an RNAqueous-Micro kit (Ambion, Austin, TX, USA) according to the manufacturer's instructions for LMD. RNA samples were quantified with an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) and the quality was confirmed with an Experion System (Bio-Rad Laboratories, Hercules, CA, USA).

Extracted total RNA was labeled with Hy5 using the miRCURY LNA™ microRNA Hy5 Power labelling kit (Exiqon, Vedbaek, Denmark). Labeled RNAs were hybridized onto 3D-Gene Human miRNA Oligo chips (v.14 1.0.1; Toray Industries, Tokyo, Japan). The annotation and oligonucleotide sequences of the probes were confirmed in the miRBase miRNA database Release 14 (http://microrna.sanger.ac.uk/sequences/). After stringent washes, fluorescent signals were scanned with the ScanArray Lite Scanner (Perkin Elmer, Waltham, MA, USA) and analyzed with GenePix Pro software (Molecular Devices, Sunnyvale, CA, USA).

Raw data were normalized by subtracting the mean intensity of the background signal, as determined from the signal intensities of all blank spots, with 95% confidence intervals. Signal intensities >2 standard deviations (SD) of the background signal intensity were considered to be valid. Relative expression of a given miRNA was calculated by comparing the signal intensities of the averaged valid spots with their mean value throughout the microarray experiments.

The data were normalized, and the significantly differentially expressed miRNAs were obtained by comparing cMCL with aMCL using a t-test (p-value <0.05). A heat map of expression data from the selected miRNAs was generated with MeV software (www.tm4.org) (17). Ingenuity Pathway Analysis (IPA6.0; Ingenuity Systems, Redwood, CA, USA; www.ingenuity.com) was used to identify miRNA interaction with genes.

Locked nucleic acid (LNA)-modified probes labeled with DIG (mercury-LNA detection probe) were obtained from Exiqon. The probe sequences were as follows: miR-15b, 5′-TGTAAACCATGATGTGCTGCTA-3′ (5′-DIG and 3′-DIG); Scramble-miR used for negative control, 5′-GTGTAACACGTCTATACGCCCA-3′ (5′-DIG); and U6 snRNA for positive control, 5′-CACGAATTTGCGTG TCATCCTT-3′ (5′-DIG). FFPE tissues section, 6-μm thin adhered to glass slides were deparaffinized in two consecutive xylene baths for 15 min each, followed by 5 min each in serial dilutions of ethanol (100, 95 and 70%) and were washed with PBS. Slides were then digested with Histo/Zyme (Diagnostic Biosystems, Pleasanton, CA, USA) at room temperature for 10 min, washed twice with PBS, fixed with 4% paraformaldehyde, and rinsed in PBS. For the hybridization step, the IsHyb In Situ Hybridization (ISH) kit (Biochain Institute Inc., Hayward, CA, USA) was used according to the manufacturer's instructions. Briefly, DIG-labeled LNA probes were denatured by heating to 90°C for 4 min and diluted to 10–100 nM in the IsHyb kit hybridization solution. Slides were hybridized overnight at 50°C in a Dako hybridizer (Dako, Glostrup, Denmark). After hybridization, slides were washed in pre-heated saline-sodium citrate (SSC) buffers at 50°C: in 2X SSC for 10 min, 1X SSC for 10 min, and 0.2X SSC for 5 min. Then, slides were placed in 0.2X SSC at 37°C for 10 min. Slides were incubated in 1x IsHyb kit blocking solution for 60 min at room temperature. Alkaline phosphatase (AP)-conjugated anti-DIG from the IsHyb kit was diluted 1:100 in PBS and applied to slides for 3 h at room temperature, followed by three washes in PBS. After two 5-min washes with 1X AP buffer, slides were incubated with nitro-blue tetrazolium chloride/5-Bromo-4-Chloro-3′-In-dolylphosphatase p-Toluidine salt (NBT/BCIP) solution from the IsHyb kit, in the dark, overnight at room temperature to develop the dark-blue NBT-formazan precipitate. Images were converted to grayscale and analyzed using the image processing and analysis software Multi Gauge ver. 3.0 (FujiFilm, Tokyo, Japan). Formalin-fixed paraffin-embedded MCL tissue samples (n=31; 19 cMCL, 3 iMCL and 9 aMCL) and non-tumorous lymph node (LN) (n=8) were analyzed.

The MCL FFPE sections (19 cMCL, 3 iMCL and 9 aMCL) used for in situ hybridization were stained with haematoxylineosin. The immunohistochemical staining of MIB1 (Ki-67: DakoCytomation, Glostrup, Denmark) was performed. Expression was scored on a four-point scale as follows: grade 0, 0–10% of the tumor cells stained; grade 1, 10–40% stained; grade 2, 40–70% stained; grade 3, >70% stained. Clinical and pathological findings for different groups were compared using the Student's t-test and the χ² test. Values were considered significant at p<0.05. The immunohistochemical analyses of p53 (mouse MAb clone DO-7: Dako Cytomation, Glostrup, Denmark) and c-myc (rabbit MAb clone Y69: Abcam, Cambridge, UK) were carried out.

We performed global miRNA expression analysis for cMCL (n=5) and aMCL (n=5) using the 3D-Gene Human miRNA Oligo chip (v.14 1.0.1; Toray) (Table I). The results showed that 7 microRNAs are significantly differentially expressed (Fig. 1 and Table III, p<0.05, t-test) between these two MCL forms. Comparing aMCL to cMCL it was observed that miR-130b, miR-15b, and miR-107 were significantly upregulated, while miR-92a-1*, miR-509-5p, miR-519e*, and miR-562 were downregulated (Table III).

Among the miRNAs identified by these miRNA microarray analyses, we focused on a miRNA which had fold change of >2 with mean signal intensity of >100. Only miR-15b (FC: 2.37, Mean signal intensity; cMCL: 292, aMCL: 690.8) fitted these criteria. This analysis raises the possibility that miR-15b plays an important role in the aggressive transformation from cMCL to aMCL.

To validate the results of global miRNA microarray analysis, we performed in situ hybridization using a mercury-LNA detection probe (Exiqon) to assess miR-15b expression. In situ hybridization was carried out for MCL Case 132, the intermediate MCL (iMCL) used for miRNA array analysis. Additionally, cMCL (19 cases), aMCL (8 cases), iMCL (3 cases), and reactive LN (8 cases) samples were also analyzed (Table II). Staining intensity of miR-15b was evaluated using image analysis software (Multi Gauge version 3.0). Fig. 2A shows a representative example of miR-15b staining for iMCL (MCL Case 132). This case exhibits morphological features of both classical and aggressive forms in the same tissue. Comparison of the two morphological forms, miR-15b intensity in the aggressive form was 2.1-fold higher than in the classical form (Fig. 1A), which correlated well with the result of miRNA array analysis (2.32-fold change, Table III). This finding suggests that miR-15b expression is significantly important in aggressive MCL.

Of note, in the reactive lymph node, miR-15b expression was higher in the germinal center; only very few cells of the mantle zone, the cells affected by MCL, expressed miR-15b (Fig. 2B).

MIB-1 labeling index correlates with the aggressiveness of MCL, and is included in MIPI. Therefore, we evaluated a correlation between miR-15b expression and MIB-1 index. The result indicated a significant positive correlation between the two (ρ=1.1005, R2=0.6287) (Fig. 3). No correlations were seen with p53 or c-myc.

In our previous study, gene expression analysis comparing cMCL and aMCL demonstrated that CDK1 expression was associated with a shift from cMCL to aMCL (10). In addition, ingenuity pathway analysis demonstrated that a connection between expression of CDK1 and miR-15b exists (Fig. 1B). These results suggest that miR-15b expression is more important in aMCL than cMCL.